The Semantics-Pragmatics Controversy 9783110333411, 9783110333374

Citation preview

Kristin Börjesson The Semantics-Pragmatics Controversy

Language, Context, and Cognition

Edited by Anita Steube Editorial Board Kai Alter, Ulrike Demske, Ewald Lang†, Rosemarie Lühr, Thomas Pechmann and Richard Wiese

Volume 14

Kristin Börjesson

The SemanticsPragmatics Controversy

DE GRUYTER

ISBN 978-3-11-033337-4 e-ISBN 978-3-11-033341-1 ISSN 1866-8313 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.dnb.de. © 2014 Walter de Gruyter GmbH, Berlin/Boston Printing: Hubert & Co. GmbH & Co. KG, Göttingen ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Acknowledgements This book grew out of my doctoral thesis, during the writing of which I was supported in different ways by various people that I would like to thank. First of all, my thanks go to Johannes Dölling for suggesting the topic to me and supervising my thesis. I have profited greatly from his assiduous comments and the various suggestions he made. I would also like to thank Anita Steube and Robyn Carston who kindly agreed to act as examiners of the thesis. It was due to Anita Steube’s initiative that I was enabled to turn my thesis into the actual book within the Language, Context and Cognition series. I am grateful for that. Robyn Carston provided a range of valuable comments on my thesis, which lead me to review and integrate some more recent (mostly relevance-theoretic) literature on the subjects in question. I started out working on my thesis in the Graduate School Universality and Diversity: Linguistic Structures and Processes at the University of Leipzig. I would like to thank the people involved in the Graduate School for making the three years I spent there a memorable time, especially my colleagues Christiane Kaden, Antje Lahne, Matthias Irmer, Susann Zschernitz, Sonja Fleischhauer, Barbara Sonnenhauser, Petra Augurzky, Andreas Bulk, Olav Müller-Reichau, Torsten Andreas, Petr Biskup, Helena Trompelt and Stefan Sudhoff. I continued to develop my thesis while working first at the Institute for German at the TU Braunschweig and later at the Institute for English and American Studies at the FAU Erlangen-Nürnberg. I would like to thank Martin Neef (TU Braunschweig) and Thomas Herbst as well as Brigitta Mittmann (both FAU Erlangen-Nürnberg) for making me feel at home at these two institutions, despite the amount of travelling involved. The final phase of this thesis I was able to concentrate on back in Leipzig, for which possibility I am very much indebted to Gereon Müller. Finally, I want to thank my mother Christine for the continuous support she gave, not only during the time it took me to write my thesis. Especially the last year, working on the thesis would not have been possible if it had not been for her looking after her (then) youngest grandson. At the risk of sounding trite, last but not least, my thanks go to Christian for supporting me and always keenly providing examples of how literal/non-literal meaning and various types of ambiguity are exploited, especially for humorous purposes.

Contents 1 1.1 1.2 1.3

Introduction 1 The Standard Notions and Their Problems 7 Aim of the Book 8 Plan of the book

2 2.1 2.1.1 2.1.2 2.1.3 2.2 2.2.1 2.2.2 2.3 2.3.1 2.3.2 2.3.2.1 2.3.2.2 2.3.3 2.3.3.1 2.3.3.2 2.3.3.3 2.3.3.4 2.3.3.5 2.3.3.6 2.3.3.7 2.4 2.4.1 2.4.2 2.5 2.6

Against the Standard Notions of Literal Meaning and Non-literal Meaning 11 11 Literal Meaning and Context-Independence 12 Literal Meaning as Compositional Meaning? 17 Literal Meaning as Context-Independent? 23 Literal Meaning as Primary to Non-literal Meaning? 28 Non-literal Meaning and Conventionality 28 Empirical Evidence 34 Theoretical Considerations 42 Consequences for Lexical Meaning 43 Problematic Data 49 Approaches to Meaning in the Lexicon 50 The Maximalist Approach 54 The Intermediate Approach 61 Semantic Underspecification in the Lexicon 61 The Minimalist Approach 66 Ruhl’s monosemic approach 67 A Cognitive Approach 72 Underspecification and Conventionality 74 Underspecification and Semantic Relations 75 More Underspecification in the Lexicon 78 Underspecification of Semantic Composition 80 Empirical Investigations of Aspects of Semantics 80 Polysemy vs. Underspecification in the Lexicon Empirical Evidence for Semantic vs. Pragmatic Processing 91 Why the Standard Notions? 95 Summary

3 3.1 3.1.1 3.1.2

Utterance Meaning and the Literal/Non-literal Distinction Levels of Meaning 100 100 Grice’s Four Types of Meaning 105 Bierwisch’s Three Levels of Meaning

1

86

99

viii

3.1.3 3.2 3.2.1 3.2.1.1 3.2.1.2 3.2.2 3.2.3 3.2.4 3.3 4 4.1 4.1.1 4.1.1.1 4.1.1.2 4.1.1.3 4.1.1.4 4.1.1.5 4.1.1.6 4.1.2 4.1.2.1 4.1.2.2 4.1.2.3 4.1.2.4 4.1.2.5 4.1.3 4.1.3.1 4.1.3.2 4.1.3.3 4.1.4 4.2 4.2.1 4.2.2 4.2.3 4.3

Contents

Summary 110 The Problem of Characterising the Level of Utterance 112 Meaning 114 Explicit/Implicit Meaning 114 Explicatures 123 Implicitures 127 Unarticulated Constituents vs. Hidden Indexicals 138 Minimal Semantic Content and Full Propositionality 147 Minimal Proposition vs. Proposition Expressed 154 Summary Utterance Meaning and Communicative Sense – Two Levels or One? 159 161 Problematic Phenomena 161 Metaphor 161 Traditional Characterisation and its Problems 163 Metaphor and The Similarity of Various Types of Meaning 166 Metaphor and Attributive Categories 168 Empirical Results Concerning Metaphor Interpretation 172 Formal approaches to metaphor interpretation 179 Summary 181 Irony 181 Traditional Characterisation and its Problems 182 Irony as echoic interpretive use 185 Irony as a Form of Indirect Negation 188 Empirical Results Concerning Irony Interpretation 195 Summary 196 Conversational Implicatures Generalised vs. Particularised Conversational Implicature – 196 Theoretical Approaches 207 (Mostly) Empirical Evidence Concerning GCIs 216 Summary 217 Speech Acts 222 Differentiating What is Said from What is Meant 224 What is Said/What is Meant and Indirect Speech Reports 229 Primary vs. Secondary Pragmatic Processes What is Said/What is Meant and Distinct Knowledge 234 Systems 242 Summary

Contents

5

ix

5.3

Varieties of Meaning, Context and the Semantics/ 246 Pragmatics Distinction Towards an Alternative Characterisation of (Non-)Literal 246 Meaning 248 Literal Meaning and Types of Non-literal Meaning 255 Literal Meaning as ‘Minimal Meaning’ 261 Nature of the Processes Determining (Non)-Literal Meaning 268 (Non-)Literal Meaning as (Non-)Basic Meaning 275 The Nature of Context in Utterance Interpretation 276 Context and the Interpretation of Implicit Meaning Aspects 276 Free Enrichment and Implicit Meaning Aspects Discourse Interpretation and Information from Conceptual 280 Frames 284 Free Enrichment and Information from Conceptual Frames 292 Consequences Context, Semantic Interpretation and the Semantics/ Pragmatics 293 Distinction 304 Summary

6

Summary

5.1 5.1.1 5.1.2 5.1.3 5.1.4 5.2 5.2.1 5.2.1.1 5.2.1.2 5.2.1.3 5.2.1.4 5.2.2

List of Figures Bibliography Index

325

307 311 313

1 Introduction 1.1 The Standard Notions of Literal Meaning and Non-literal Meaning and Their Problems One of the major issues in investigating the relation of language and meaning is the question of how to characterise and draw the line between what traditionally are called semantics and pragmatics. In describing what they take to be the characteristics of one or the other system, linguists often make use of the terms literal meaning and non-literal meaning. For example, Lyons (1987) lists a number of propositions used in the differentiation of semantics from pragmatics, amongst which is the following: ‘...that semantics deals with literal, and pragmatics with non-literal, meaning...’ (ibid., p. 157). Similarly, Cole (1981, p. xi) states that semantics ‘...is involved in the determination of conventional (or literal) meaning...’, whereas pragmatics is concerned with ‘...the determination of nonconventional (or nonliteral) meaning...’ and Kadmon (2001, p. 3) writes ‘...I think that roughly, semantics only covers “literal meaning.” Pragmatics has to do with language use, and with “going beyond the literal meaning.”’. More recently, Recanati (2004, p. 3) summarised (and criticised) the standard view on the division of labour between semantics and pragmatics, starting as follows. ‘Semantics deals with the literal meaning of words and sentences as determined by the rules of the language, while pragmatics deals with what the users of the language mean by their utterances of words and sentences’. For such a characterisation of semantics and pragmatics to be useful, one has to know how the kinds of meaning the terms literal meaning and non-literal meaning refer to are characterised. This is problematic in so far as one usually does not find such characterisations in the literature. Generally, it rather seems that the two terms literal meaning and non-literal meaning are treated as denoting basic kinds of meaning that are intuitively clear and as such need no further description. The pair of terms literal meaning/non-literal meaning actually is only one of quite a number of dichotomies used in the characterisation of semantics and pragmatics. Thus, the two systems are often characterised in terms of the differentiation between conventional vs. non-conventional meaning, as, e.g. in the quote from Cole (1981) given above. See also again Lyons (1987), who lists the proposition ‘...that semantics has to do with conventional, and pragmatics with the non-conventional, aspects of meaning...’ (ibid., p. 157). Another important pair of terms traditionally used is context-independent vs. context-dependent meaning. Thus, Lyons (1987, p. 157) states ‘...that semantics deals with contextindependent, and pragmatics with context-dependent, meaning’. More specifi-

2

Introduction

cally, Katz (1977) introduces the notion of the ‘anonymous letter situation’ to characterise the kind of meaning captured by semantics in contrast to pragmatics. [I] draw the theoretical line between semantic interpretation and pragmatic interpretation by taking the semantic component to properly represent only those aspects of the meaning of the sentence that an ideal speaker-hearer of the language would know in an anonymous letter situation, ... [where there is] no clue whatever about the motive, circumstances of transmission, or any other factor relevant to understanding the sentence on the basis of its context of utterance. (Ibid., p. 14)

In addition, semantics is also characterised as dealing with those aspects of meaning that expressions have, independent of their use. In contrast, pragmatics is understood as dealing with those aspects of meaning that are determined by the actual use of language. Thus, compare again Lyons (1987) who mentions the idea that ‘...semantics has to do with meaning, and pragmatics with use...’ (ibid., p. 157). Accordingly, one finds uses of the terms literal meaning and non-literal meaning which pick up on this view of the difference between semantic and pragmatic meaning. For instance, Bach (2001a) writes Words do not have nonliteral meanings [...], but they can be used in nonliteral ways. [...] In familiar cases, such as metaphor and metonymy, particular expressions are used nonliterally. [...] But there is a different phenomenon which I call “sentence nonliterality,” [...] Here a whole sentence is used nonliterally, without any of its constituent expressions being so used. (Ibid., p. 249, my emphasis)

Thus, whereas literal meaning is a feature that expressions are said to have, the non-literal meaning of an expression results from the particular use of that expression. To summarise the standard understanding of semantics and pragmatics: whereas the former is characterised as dealing with literal, conventional and context-independent meaning, the latter deals with non-literal, non-conventional and context-dependent meaning. Using the dichotomies in this characterisation suggests that there is a correspondence between literal, conventional and contextindependent meaning, on the one hand, and non-literal, non-conventional and context-dependent meaning on the other.1 In other words, the fact that the two terms literal meaning and non-literal meaning are used amongst others in a dichotomous characterisation of semantics and pragmatics suggests that these other terms also may be used in characterising literal meaning and non-literal mean-

1 From the quotes given above, this is especially apparent in Cole’s, who uses the terms literal and non-literal as synonymous to conventional and non-conventional, respectively.

The Standard Notions and Their Problems

3

ing as such. In fact, this implicit assumption has led to what might be called the standard notions of literal meaning and non-literal meaning which are summarised in what follows. Literal meaning, on the one hand, is assumed to be conventionalised, that is, it does not take any special interpretation effort to arrive at it. The literal meaning of simple expressions is listed in their lexical entries; the literal meaning of complex expressions is the result of a principled combination of the literal meanings of their parts. Thus, both the literal meaning of simple as well as complex expressions is characterised by the fact that it is context-independent. Non-literal meaning, on the other hand, is assumed to be non-conventionalised, thus, it does take a special interpretation effort to arrive at it. Intuitively, it is considered as deviating from some more basic (literal) meaning in a fairly special way. Moreover, in contrast to literal meaning, non-literal meaning crucially is taken to be context-dependent. Overall, the term non-literal meaning is used to differentiate from literal meaning a kind of meaning that is derived from the latter and, in a sense, has a secondary status. Therefore, it is traditionally assumed that in terms of the enfolding of the interpretation process, the literal meaning of an expression is processed first, whereas any potential non-literal meanings are processed afterwards and only if the literal interpretation does not fit the given context. However, as the extensive debate concerning the proper demarcation of semantics from pragmatics – especially of the last 15 years (cp. Carston 1999, Turner 1999, Dölling 2001, Bianchi 2004, Borg 2004b, Cappelen and Lepore 2005, Horn 2006, Dölling and Zybatow 2007, Carston 2009, Frisson 2009, Recanati 2010, Borg 2012, Carston and Hall 2012, etc.) – shows: not only is it unclear whether the standard notions of literal meaning and non-literal meaning actually are useful in the characterisation of distinctive kinds of meaning aspects, what is even more problematic is the fact that they are based on an understanding of semantics and pragmatics that has come under increasing criticism. In particular, the question of whether semantics should be taken to be differentiated from pragmatics by the property of context-(in)dependence of meaning has been – and still is – heavily discussed. This has become a pressing question since, in addition to the assumption that it deals with context-independent meaning, semantics has also traditionally been characterised as determining the proposition expressed by a sentence. However – and this is implicit already in Grice (1975, 1989)’s characterisation of the two levels of meaning what is said and what is meant – it can be argued that semantics alone actually does not determine the proposition expressed by a sentence. Thus, although Grice characterised the level of what is said as ‘...closely related to the conventional meaning of the words (the sentence) [the speaker] has uttered’ (Grice 1989, p. 25), he also recognised that for a sentence to express a determinate proposition at all, indexicals have to get fixed

4

Introduction

and ambiguities and references resolved. Grice himself did not explicitly call the processes that lead to such specifications of meaning either semantic or pragmatic. However, in the discussion of the two levels of meaning that followed, some authors have claimed that what is said – or at least a level of meaning very similar to it – actually IS the semantic content of an utterance (and as such is determined by a semantics component that does allow the consideration of contextual information after all), whereas others have argued that it is a level of meaning that has already gone beyond the purely semantically determined content (thus keeping to the traditional view of semantics as independent of contextual information).2 Be that as it may (for now), the important point to note is that – considering that traditionally semantics is in fact characterised by both the properties of dealing with context-independent meaning as well as determining the propositions expressed by sentences and considering that the characterisation of literal meaning derives from that of semantics – maybe it actually is the latter mentioned property of semantics that the term literal meaning should be taken to relate to. In other words, maybe it is not the context-independent meaning that is literal, but rather the proposition expressed by a sentence. If the latter is the case, then literal meaning would in fact not be context-independent. Actually, Korta and Perry (2008) claim that ‘[w]hat is said has been widely identified with the literal content of the utterance...’ and looking at the quotations below, where the term literal is indeed used to refer to a context-dependent level of meaning (roughly: Grice’s what is said), this claim is corroborated. So, for instance, Carston (2007, p. 21) speaks of the ‘...literal meaning of [a speaker’s] utterance’. Similarly, Recanati (1995, p. 2) refers to ‘...the literal interpretation of an utterance (the proposition literally expressed by that utterance)...’ and Sag (1981, p. 274-5) speaks of the ‘...propositional content of an utterance (i.e., its literal meaning)...’. It should be noted that although these authors reject the standard characterisation of semantics and pragmatics and they use the term literal meaning in a non-standard understanding, they only do the latter implicitly. That is, these authors do not explicitly say anything new concerning the properties that characterise literal meaning and non-literal meaning, respectively. In fact, Bierwisch (1979, 1983, 1997) is the only exception here in that he explicitly uses the term literal meaning with respect to a context-dependent level of meaning he calls utterance meaning, which

2 In fact a third possibility has been proposed, namely that semantics DOES indeed have both the properties of dealing with context-independent meaning only and for sentences determining the propositions expressed by them. However, the thus determined level of meaning is taken to be distinct from Grice’s level of what is said. For a detailed discussion of the various different approaches to the semantics/pragmatics distinction, see chapter 3.

The Standard Notions and Their Problems

5

is quite similar to Grice’s what is said. Thus, he says of an expression’s utterance meaning that it may or may not correspond to the literal meaning this expression has in that particular utterance context. However, he does not give any details as to how this particular type of meaning is determined or differentiated from others. Generally, what the quotations given so far show is that the term literal meaning is not only used with respect to a context-independent level of meaning. Rather, and as Bezuidenhout and Cutting (2002, p. 435) note, ‘[t]he phrases “literal meaning” or “literal interpretation” have been used to cover both the literal meaning of a sentence and what is said by the utterance of a sentence in a context’. In other words, the term literal is used to refer to quite different types of meaning levels. In fact, the pair of terms literal and non-literal is even used in the characterisation of so-called indirect speech acts – usually taken to belong to the pragmatically determined level of meaning what is meant – which have been analysed as being associated with two illocutionary forces, where one is the primary and at the same time non-literal and indirect speech act and the other is the secondary and at the same time literal and direct speech act. The literal speech act is the illocutionary force taken to be conventionally associated with the particular sentence-type used for the expression of some particular speech act, whereas the non-literal speech act is the act actually intended by the speaker. Thus, it seems the two terms literal meaning and non-literal meaning are (mostly) used based only on intuitions we have concerning the nature of the relation between particular types of meaning aspects rather than on an identification of determinate and contrasting sets of properties those types of meaning aspects can be shown to exhibit. This becomes apparent when looking in more detail at the properties used in the standard characterisations of the two terms and the phenomena intended to be picked out by them, where it turns out that the phenomena do not all show the properties suggested by the standard characterisation. Thus, there is an argument to be made that literal meaning should not be viewed as context-independent (as we already saw), always conventional and always primary in interpretation. Similarly for non-literal meaning, one does not necessarily have to assume that it is always non-conventional and secondary in interpretation.3 What complicates matters even further is the fact that the notions traditionally used in the standard characterisations of literal meaning and non-literal meaning – and of semantics and pragmatics, of course – such as (non)-conventionality and context-(in)dependence, actually are problematic themselves. Thus, the use

3 See chapter 2 for detailed arguments supporting such a view.

6

Introduction

of the pair of terms conventional vs. non-conventional as exemplified above suggests that conventionality is an all-or-nothing property. However, as is suggested by the results of various experiments investigating the nature of the interpretation process on the one hand (cf. Giora 1997, 1999, Gibbs 2002), as well as by theoretical considerations within the field of historical semantics on the other (cf. Busse 1991), this view is an oversimplification of the facts. Similarly – and as mentioned above already – not all approaches that are characterised as essentially semantic by their proponents necessarily share the view that what semantics deals with is context-independent meaning only (cf. Sag 1981, Borg 2004b, Cappelen and Lepore 2005). Having said that, it should be noted that there is no single concept of what constitutes a context, but rather several. Thus, even if different authors claim that semantics is context-dependent after all, actually they may not agree on which processes exactly are involved in determining semantic meaning or on the kind of contextual information that plays a role in that determination. Generally, it is questionable whether the terms literal meaning and non-literal meaning can be characterised and differentiated in terms of the dichotomies traditionally used. The same concern holds for the characterisations of semantics and pragmatics from which – as we saw – that of the terms literal meaning and non-literal meaning derives. Yet another problem is that with only the standard notions of literal meaning and non-literal meaning to rely on, it is no trivial question to ask how these two meaning aspects are related to other kinds of meaning aspects identified in the individual approaches, such as e.g., explicit/implicit meaning aspects of an utterance due to free enrichment, so-called ad-hoc concepts or conversational implicatures. According to the standard characterisation, they should all be cases of non-literal meaning as all of them are context-dependent. However, it can be argued that this is stretching the notion of non-literal meaning a bit too far, especially as it involves the grouping together of meaning aspects which otherwise are very different in nature.4 To summarise the main points made so far: although there exist some standard characterisations of the terms literal meaning and non-literal meaning, they are not always used in accord with these characterisations, indicating that the latter are not appropriate. Moreover, although there exist alternative approaches to the standard differentiation of semantics from pragmatics, these approaches largely remain silent about whether – and if so, how – the standard notions of literal meaning and non-literal meaning should be revised. In fact, as in the traditional literature, if the terms literal meaning and non-literal meaning are used, this

4 See chapters 3 and 4 for further details.

Aim of the Book

7

is done under the assumption that it is clear what they refer to and how they can be differentiated from other types of meaning aspects. However, as already mentioned above, this is not at all clear. The only notable exception here is Bierwisch, who clearly assumes of literal meaning that it is context-dependent but does not explain why he makes that assumption. Moreover, although Bierwisch characterises literal meaning as a particular type of utterance meaning, he does not say anything either as to how this particular type of meaning is determined. Thus, the present book actually ties up to Bierwisch’s assumption concerning the nature of literal meaning but goes further in that it gives reasons for why this assumption is reasonble to make and explicitly asks how literal meaning is determined and how it is differentiated from non-literal meaning, on the one hand, as well as other types of meaning aspects on the other. More generally, it shows in detail why the standard notions of literal meaning and non-literal meaning are inadequate.

1.2 Aim of the Book The problems sketched in the last section led me to the formulation of the three questions below, which I aim to answer in the present book. 1. 2.

3.

What is it that makes the standard notions of literal meaning and non-literal meaning inadequate and thus in need of revision? What exactly are the properties that characterise and differentiate literal meaning and non-literal meaning and how are these particular types of meaning related to other types of meaning identified in the semantics/pragmatics literature (e.g., conversational implicature, implicit meaning aspects)? By which criteria should semantics and pragmatics be characterised and differentiated, if not by the dichotomies traditionally used and under the assumption that the two systems are involved in the determination of (at least) three distinct meaning levels in interpretation?

To answer the first question, I will look at the individual properties standardly assumed to be exhibited by literal meaning and non-literal meaning and show that they cannot all simultaneously hold. More specifically, I will give arguments that actually both literal meaning as well as non-literal meaning are context-dependent and that they are not differentiated by conventionality of meaning. I will further argue that the two terms – as well as the dichotomies mentioned above in general – cannot be used in the characterisation of semantics and pragmatics. In order to answer the second question – but also as a preliminary for answering the third – I will review and compare different, currently prominent ap-

8

Introduction

proaches to utterance interpretation as well as consider empirical data on various relevant phenomena. The focus will be on the identification of the levels of meaning assumed in the individual approaches and how these are characterised, as well as on the respective characterisations of the particular types of meaning aspects and interpretation processes identified. Based on the discussion of the different approaches to utterance interpretation and the various aspects of meaning as well as on a defence of the appropriateness of differentiating two context-dependent levels of meaning, I will finally formulate my answer to the third question. Generally, the primary aim of this work is not so much to offer a ‘new’ model of utterance interpretation that integrates semantics and pragmatics. Rather, I have compared existent theories of utterance interpretation as to the basic notions they make use of and how these relate to semantics or pragmatics.

1.3 Plan of the book The book is structured as follows. In chapter 2, I will argue against the standard notions of literal meaning and non-literal meaning. In particular, I will argue against the traditional characterisation of literal meaning and non-literal meaning, according to which the former is taken to be context-independent and the latter non-conventional. Having established that literal meaning does not necessarily have to be taken to be context-independent and as such semantic in nature, I will discuss the consequences this view has for the nature of lexical meaning. After reviewing a number of different types of approaches to lexical meaning, I will argue for a view that assumes a high degree of underspecification of lexical meaning. Generally, in the discussions in chapter 1, I will consider both theoretical viewpoints as well as empirical data. In particular, one section is dedicated to empirical studies on aspects of the semantics component, namely that lexical meaning is characterised by underspecification and that, generally, semantic processes of meaning construction should be differentiated from pragmatically based plausibility checks. In the last part of chapter 1, I will try to answer the question of why the standard notions of literal meaning and non-literal meaning came to be assumed in the first place. Here, the idea of stereotypical interpretations of linguistic expressions presented ‘out of context’ will be considered. Having argued against the standard notions in chapter 2, and more specifically, having argued for viewing literal meaning, similarly to non-literal meaning as essentially context-dependent as well, chapter 3 is dedicated to looking in detail at the first context-dependent level of meaning called what is said by Grice, to see how this has been characterised subsequently and to identify the process-

Plan of the book

9

es potentially involved in determining literal meaning at this level of meaning. I will start with Grice’s differentiation of four different types of meaning and relate them to the two levels of meaning Grice introduced: what is said and what is meant. Following that, I will present Bierwisch’s threefold differentiation of levels of meaning, based on the different knowledge systems made use of in their determination. In the second part of chapter 3, I will discuss a range of approaches that give alternative characterisations for Grice’s level of what is said. The overall aim is to identify the different processes at work in determining what is said, how these processes are characterised and which types of meaning aspects can be found at this level of meaning (appart from potentially literal or non-literal meaning). At the same time, the various approaches discussed also all offer slightly different views on the nature of the semantics and pragmatics components and how they interact in the process of utterance interpretation. While the greater part of chapter 3 is taken up by theoretical considerations, towards the end of that chapter a few empirical results will also be discussed. Chapter 4, then, is concerned, on the one hand, with phenomena traditionally assumed to arise at Grice’s level of meaning what is meant, and, on the other hand, with the more basic question of whether a differentiation of two contextdependent levels of meaning what is said and what is meant actually is necessary/possible. Thus, in the first part of chapter 4, alternative approaches to the phenomena of metaphor, irony, (primarily generalised) conversational implicature and (primarily indirect) speech acts will be reviewed as well as empirical results considered that test the predictions following from the individual approaches. Here, the aim is to establish, on the one hand, how these different meaning aspects are determined and, on the other hand, which of the phenomena actually can be usefully considered as non-literal. More generally, the question is addressed at which level of meaning (i.e. what is said or what is meant) the individual phenomena should be taken to arise. In the second part of chapter 4, various arguments will be presented for and against differentiating the two levels what is said and what is meant from one another. I hope to make clear that such a differentiation is useful and necessary, although it might be difficult to decide on the criteria to be used in this differentiation. Chapter 5, finally, turns back to the basic question that chapter 2 ends with, namely how literal meaning and non-literal meaning actually should be characterised if one wants to capture the various uses the two terms are put to. I will start out with two alternative characterisations of what literal meaning and nonliteral meaning should be taken to be, before presenting my own characterisation, based on the discussion in the preceding chapters. As a preliminary for my characterisation, I will review the various processes identified in the preceding chapters as involved in the overall interpretation of utterances. The main consequence

10

Introduction

drawn from my characterisation of literal meaning and non-literal meaning will be that these two notions actually cannot be used in the characterisation of the distinction between semantics and pragmatics, if the former, in contrast to the latter, is essentially taken to be context-independent. The last part of chapter 5 will take up exactly this point, namely the nature of contextual information in utterance interpretation and whether the notion of context-(in)dependence actually is useful in differentiating between semantics and pragmatics. Thus, I will first offer a proposal concerning the nature of the contextual information the process of free enrichment makes use of. Free enrichment is one of the processes assumed to contribute to the level of utterance meaning and crucially is taken to depend on a consideration of potential speaker intentions for its operation. I will show that this assumption is not necessary, allowing for a differentiation between the level of utterance meaning and communicative sense based on the assumption that processes leading to the former do not involve reasoning concerning the speaker’s potential intentions in making the utterance she did, whereas the processes leading to the latter level of meaning do. In the final section of chapter 5, I will turn back to the characterisation of the semantics/pragmatics distinction and after discussing a number of views on that characterisation present my own.

2 Against the Standard Notions of Literal Meaning and Non-literal Meaning In this chapter, I will argue against the standard notions of literal meaning and non-literal meaning described in chapter 1 (sections 2.1 and 2.2, respectively). The main aim of this chapter is to show that the dichotomies traditionally used to differentiate literal meaning from non-literal meaning either cannot in fact differentiate the two meanings (as is the case with the feature of context-(in)dependence) or are not such ‘all-or-nothing’ concepts as traditionally implied (as is the case with the property of conventionality). Generally, the arguments presented point to the crucial conclusion that literal meaning and non-literal meaning are in fact not so different from one another as traditionally assumed. Having argued against viewing literal meaning as essentially context-independent and non-literal meaning as essentially non-conventional, I will consider the consequences this has for the nature of lexical meaning (section 2.3). Moreover, I will consider empirical evidence supporting the assumption of underspecification of lexical meaning and, more generally, a distinction between distinctly semantic and pragmatic processes in interpretation (section 2.4). In addition, I will address the question of why the standard assumptions came into existence in the first place (section 2.5).

2.1 Literal Meaning and Context-Independence Traditionally, complex expressions are assumed to have literal meaning in the form of what formal semantics1 calls sentence meaning, which results from the process of semantic composition which combines the literal meanings of the simple expressions that together constitute the complex expression and which captures the proposition expressed by that sentence.2 Moreover, during interpretation, the literal meaning of a complex expression is computed first, whereas its potential non-literal meaning is computed afterwards and only if the literal meaning does not fit the given context (cf. Grice 1975; Searle 1979).

1 Note that in what follows, on the semantics side, I am primarily interested in assumptions made in the programme of formal semantics. 2 In this section, I will mainly be concerned with two of the three properties literal meaning is standardly claimed to exhibit, namely that it is context-independent and primary to non-literal meaning. Also, I will concentrate on complex expressions, leaving the discussion of the nature of lexical meaning to section 2.3.

12

Against the Standard Notions of Literal Meaning and Non-literal Meaning

Intuitively, these characterisations seem to be sound. They give a fairly general description of what we take literal meaning to be with respect to complex expressions. However, looking at each of the characteristics in more detail reveals that they are not unproblematic. Thus, it is questionable whether what we usually take to be a complex expression’s literal meaning does in fact correspond to its context-independent, compositional meaning. Put differently, the question is whether the formal semantic notion of sentence meaning can be assumed to both be the sum of the lexical meanings of the simple expressions involved as well as having a fully propositional form. Furthermore, in computing the ‘speakerintended’ non-literal meaning of an expression, it may not actually be necessary to first compute the literal meaning of the expression the speaker used as an intermediate step.

2.1.1 Literal Meaning as Compositional Meaning?

Concentrating on the traditional characterisation of the programme of formal semantics and the role of literal meaning therein reveals that, in a sense, the characterisation of literal meaning and non-literal meaning is interdependent on the characterisations of semantics and pragmatics. Thus, basically, formal semantics can be characterised as dealing with the context-independent meaning of simple and complex expressions.3 More specifically, it aims at formulating truth conditions for sentences. That is, it takes as a starting point for analysis the level of sentence meaning, mainly for two reasons. First, it seems that sentences express propositions, that is, complete thoughts, something of which it makes sense to ask whether it is true or not. Second, intuitively at least, the meaning of a sentence can be grasped without any reference to an actual utterance of that sentence and is thus context-independent. It contrasts with interpretations of a sentence that can only be derived by considering the actual context in which that sentence is uttered (e.g., cases of irony or particularised conversational implicature). Thus, sentence meaning is considered literal in the sense that its derivation is independent of contextual information. Moreover, sentence meaning also is the level from which the meanings of the individual expressions involved are derived, following the principle of compositionality. And since sentence meaning is context-independent, the meanings of the simple expressions derived from it are context-independent too. They are the lexical meanings of the expressions con-

3 See below, however, for formal semantic approaches that also take into account contextual information.

Literal Meaning and Context-Independence

13

cerned. Thus, primarily, what the term literal meaning refers to is a certain type of meaning that, intuitively, seems to differ from other types of meaning mainly by virtue of the fact that it is context-independent and fully propositional (sentence meaning). Derivatively, the term also refers to types of meaning which are not propositional, but crucially are context-independent and are derived from a full proposition via the principle of compositionality (lexical meaning). meaning

context-independent

lexical

compositional

context-dependent

...

l i t er al

...

...

nonl i t er al

Fig. 2.1: Traditional differentiation of types of meaning

So far so good. However, the characterisation of formal semantics as stated above has proven to be problematic. And, as we will see, these problems also extend to the characterisation given to the notion of literal meaning. Thus, to summarise: in its traditional form, three of formal semantics’ main assumptions are the following: a) semantics is concerned with the context-independent meaning of natural language expressions, b) for sentences, what is determined by the semantic component of a natural language grammar is the proposition expressed by that sentence and c) for simple expressions their semantics (or lexical meaning) is whatever aspects of their meaning remain constant across different uses of that expression. However, as Sag (1981) points out: a formal semantic theory which does not allow for any contextual information to be made use of in determining the proposition expressed by a sentence ‘...appears to be falsified by the mere existence of sentences containing tense morphemes or other indexical expressions.’ (Ibid, p. 274). Thus, consider the sentence in (1). (1) He went to the bank yesterday. For the sentence in example (1), it is clearly not the case, that semantics determines a truth-evaluable proposition, due to the occurrences of the contextdependent expressions he and yesterday as well as the homonymous noun bank. As is the case for all indexical expressions, the exact reference of he and yesterday differs with the contexts in which they are uttered. Thus, for such expressions se-

14

Against the Standard Notions of Literal Meaning and Non-literal Meaning

mantics only gives rules for ’where to look’ in the search for potential referents. In the case of the occurrence in a sentence of homonymous expressions such as bank, the assumption is that the process of semantic composition has to build up as many different structures for the sentence, as there are ambiguous expressions in it. Thus, for (1) to express a full proposition, the references of the occurring indexical expressions have to be fixed, that is, recourse has to be taken to the context of the utterance. Moreover, the sentence has to be disambiguated, which, again, is only possible with the help of contextual information. Even then, the sentence does not express a proposition until the reference of the NP the bank to some unique location has been fixed. What this shows is that the proposition expressed by some sentence can only actually be determined once the context in which it is uttered is taken into consideration. Thus, it seems that the semanticist cannot uphold both assumptions a) and b). If he wants to rescue assumption a), it seems he has to concede that, in fact, the semantic component does not determine the truth-evaluable proposition expressed by a sentence; if he wants to rescue the assumption in b), he has to allow for context-sensitive processes to take place during the determination of the proposition expressed by a sentence. However, formal semantic approaches exist which attempt to capture the difference between context-sensitive and context-insensitive expressions and at the same time uphold assumptions a) and b). One such approach is proposed by Kaplan (1989b). Thus, Kaplan proposes to differentiate between, in a sense, two meanings of expressions: their character and their content.4 Consider example (2). (2) a. Mary: I am hungry. b. John: I am hungry. On the one hand, the notion of character captures the intuition that Mary and John in a way have said the same thing: both used the same sentence. The notion of content, on the other hand, captures the intuition that, at the same time, Mary and John have not expressed the same idea. Kaplan’s suggestion is that a sentence’s character is a function that takes a context in order to deliver a proposition or the content of that sentence in that context. Thus, although Mary and John use the same sentence, they express different propositions: Mary says that she is hungry, whereas John says that he is hungry. This difference is due to the character of I, which can be glossed as ‘referring to the speaker or writer’. Applying I’s character to a particular context determines the actual speaker in that context,

4 Cf. Chierchia and McConnell-Ginet (2000), Braun (2010) for accessible introductions to these notions.

Literal Meaning and Context-Independence

15

i.e., the content (or intension) of I in that particular context. Having determined the proposition expressed by a sentence in a particular context, the proposition can then be evaluated with respect to a circumstance of use or possible world. Thus, the content (or intension) of a sentence in a particular context is a function from possible worlds to truth values. In Kaplan’s approach, then, context-sensitive expressions are such that their character applied to different contexts yields different contents. However, a context-sensitive expression’s content in turn is a constant function from possible worlds to extensions since regardless of the world at which the content of the expression is evaluated, it will always have the same extension. For example, the content of an expression such as I varies depending on the context in which it is used. However, once the content is determined, it stays the same for all possible worlds. In contrast, the content of hungry does not depend on the context in which it is used. It always is the property ‘being-hungry’. However, the actual extension of this predicate depends on the possible world that is assumed. That is, the set of individuals to which the predicate applies may differ across different worlds. Thus, context-insensitive expressions have varying extensions, while their characters are such that regardless of the context the respective character is applied to, the same content will be determined. In a way, for context-insensitive expressions their character and content fall together. Kaplan’s approach, thus, allows a differentiation of three levels of meaning: character, content or intension and extension. For sentences this means one can differentiate between the context-independent sentence meaning, the proposition expressed by a sentence in a context and the truth value of a sentence in a context with respect to a possible world. Thus, implementing these ideas in a modeltheoretic semantic apparatus leads to the truth of a sentence not only being determined with respect to a world and time, but also a context of utterance (cf. Sag 1981). In a way, within such an approach, both assumptions a) and b) can be maintained. That is, what is determined by the semantic component is the contextindependent meaning of a sentence and the conditions under which that sentence is true. Using the indices w, i and c, the instruction of how to determine the proposition expressed by a sentence is also given. However, it should be noted that the proposition expressed only actually is determined, once the functions are applied to a particular world, time and context. In other words, although it is possible within such an approach to formulate conditions under which a particular sentence is true, due to the indices used the sentence’s meaning thus given may be compatible with quite a number of different situations. Thus, it cannot be taken to represent the proposition expressed in a particular utterance situation.

16

Against the Standard Notions of Literal Meaning and Non-literal Meaning

Traditionally, formal semanticists have assumed that the semantic component of the language faculty determines the meaning both of simple and complex expressions and then there are a restricted number of processes (namely, resolving of reference, fixing of indexicals and disambiguation) that lead to the proposition expressed by a sentence. However, these processes are not explicitly referred to as being of a pragmatic nature. This is quite obvious in the works of Grice, who mentions the processes that lead to what he called what is said, but does not seem to consider them as pragmatic in the same sense as the processes that result in conversational implicatures (Grice 1975) (cf. figure 2.2). However, if pragmatic processes are characterised by the fact that they take into account contextual information then, surely, the processes of fixing indexicals, resolving references and disambiguation are of pragmatic nature. Semant i c s

lexical meaning s emant i cc ompos i t i on

sentence meaning ?

r ef er enc er es ol ut i on f i x i ngi ndex i c al s di s ambi guat i on basis for further pragmatic inferences c onv er s at i onal i mpl i c at ur e s peec hac t s ...

what is said

Pr agmat i c s

what is meant

Fig. 2.2: Grice’s distinction of what is said and what is meant

Thus, as Strawson (1950) noted, it is not sentences which express something of which it makes sense to ask whether it is true or false but rather the utterances of those sentences. Thus, one and the same sentence can be used to express something true at one point and something false at another. That is, regardless of whether sentences include indexical or ambiguous expressions, it is not a general property of sentences, but rather of utterances that they express propositions. If it is not sentences per se that express propositions and are truth-evaluable but rather their utterances, what exactly, then, does the concept of sentence meaning capture? This is an important question considering that formal semantics takes sentence meaning as the starting point from which to deduce the meanings of simple expressions, which presupposes that the notion of sentence meaning is clearly defined. A possible answer is to still regard both the meaning of simple as well as complex expressions, in particular sentences, as essentially context-

Literal Meaning and Context-Independence

17

independent. That is, as traditionally assumed, semantics deals with the meaning of both simple and complex expressions, where the meaning of simple expressions forms part of their lexical entries and the meaning of complex expressions is a function of the meanings of their parts and their syntactic combination. However, such a view does not claim that sentence meaning necessarily is propositional; it simply assumes that sentence meaning is context-independent.

2.1.2 Literal Meaning as Context-Independent?

But what about the correlation between sentence meaning and literal meaning suggested above? There it was stated that, apparently, literal meaning refers to a level of meaning identified as sentence meaning by traditional formal semantics and characterised as being context-independent and fully propositional. The assumption was that the notion of literal meaning mainly captures the fact that sentence meaning is context-independent, thus, with the revised characterisation of sentence meaning as ‘only’ context-independent but not necessarily fully propositional, the term literal meaning should still be applicable to that level of meaning. There are a number of considerations that go against this characterisation. Thus, recall the uses of the term literal meaning mentioned in chapter 1, where the term, on the one hand, is used to refer to a kind of context-dependent but at the same time in some sense ‘basic’ meaning and, on the other hand, is contrasted with a kind of meaning that is not only context-dependent but crucially in some sense ‘derived’ or non-basic (cf. Sag 1981, Recanati 1995, Carston 2007). As mentioned before, such a use calls into question the adequacy of characterising literal meaning as context-independent meaning. In fact, already in his (1978) paper, Searle criticised this characterisation of literal meaning. He argues that there is no such thing as a solely linguistically determined literal meaning of a complex expression. As regards sentence meaning, one cannot speak of the literal meaning of a sentence in the standard sense. As Recanati (2004) puts it, Searle holds the view of contextualism, according to which ‘...there is no level of meaning which is both (i) propositional (truth-evaluable) and (ii) minimalist, that is, unaffected by top-down factors.’ (Ibid., p. 90). Thus, Searle assumes that the expression of a determinate proposition takes place against a set of background assumptions. To illustrate his point of view, Searle uses the sentence in (3), which, taken out of context, seems to have a quite obvious literal meaning, which, following Searle (1978), I depict as in 2.3. (3) The cat is on the mat.

18

Against the Standard Notions of Literal Meaning and Non-literal Meaning

Fig. 2.3: The typical cat-on-the-mat configuration

The problem with this ‘literal sentence meaning’ is that although speakers or hearers are not necessarily aware of the fact, a number of preconditions are assumed to hold.5 To show this, Searle constructs a context of utterance for the sentence in (3), where it is questionable whether one would want to say that the sentence correctly describes the state of affairs at hand. suppose the cat and the mat are in exactly the relations depicted only they are floating freely in outer space, perhaps the Milky Way galaxy altogether. In such a situation the scene would be just as well depicted if we turn the paper on edge or upside down since there is no gravitational field relative to which one is above the other. Is the cat still on the mat? (Searle 1978, cited from Searle 1979, p. 122)

Thus, if what the meaning of a sentence does is determine a set of truth conditions, Searle argues that for most sentences this determination can only take place against specific background assumptions. These background assumptions are not part of the semantic structure of the sentence, that is, they are unarticulated. Moreover, due to possible variations in the background assumptions, the same sentence might have varying truth conditions. For any sentence, there is no fixed set of background assumptions of which it could be said that it determines that sentence’s literal meaning. To illustrate this fact, Searle construes a context of utterance for (3), in which it could be used to truthfully describe a situation such as depicted in figure 2.4.6

5 Note that Searle is not referring to the fact that the sentence in (3) additionally contains indexical elements. That is another matter. 6 This is Searle’s context: ‘The mat is in its stiff angled position, as in [figure 2.4], and it is part of a row of objects similarly sticking up at odd angles - a board, a fence post, an iron rod, etc. These facts are known to both speaker and hearer. The cat jumps from one of these objects to another. It is pretty obvious what the correct answer to the question “Where is the cat?” should be when the cat is in the attitude depicted in [figure 2.4]: The cat is on the mat.’ (Searle 1978, cited from Searle 1979, p. 125).

Literal Meaning and Context-Independence

19

Fig. 2.4: A rather unusual cat-on-the-mat configuration

A further example for the fact that the literal meaning of a sentence depends on background assumptions can be found in Searle (1980). Searle gives a number of sentences containing the verb to cut; here are the first five. (4) a. Bill cut the grass. b. The barber cut Tom’s hair. c. Sally cut the cake. d. I just cut my skin. e. The tailor cut the cloth. As Searle notes, in each of the example sentences in (4) cut occurs in its literal meaning. There is nothing in these sentences as such that would lead one to interpret them as metaphorical or figurative. However, although cut occurs in its literal meaning, the situations that it is used to describe differ conceptually. Thus, although cut is used in its literal meaning, for the different sentences in (4), it determines different truth conditions. This can be seen if one considers what it would mean to obey an order of cutting something. Searle puts it as follows. If someone tells me to cut the grass and I rush out and stab it with a knife, or if I am ordered to cut the cake and I run over it with a lawnmower, in each case I will have failed to obey the order. That is not what the speaker meant by his literal and serious utterance of the sentence. (Searle 1980, p. 223)

Thus, again, in the examples in (4), the literal meaning of the individual sentences (and of the word cut) is determined against a set of background assumptions, namely what we know about lawns and cakes and so on and what are usual actions in which we involve with regard to those ‘things’. Furthermore, in his discussion on the cut examples, Searle points out that it is not sufficient to assume that the different readings of cut – its different literal meanings are due to some intrasentential interaction between the verb and its internal argument. That is, he argues against the view according to which cut together with the respective argument determines that cut in ‘cut the grass’ will

20

Against the Standard Notions of Literal Meaning and Non-literal Meaning

receive a different interpretation from the one it receives in ‘cut the cake’. His reasoning is that it is possible to ‘...imagine circumstances in which “cut” in “cut the grass” would have the same interpretation it has in “cut the cake”...’ (Searle 1980, p. 224). Suppose you and I run a sod farm where we sell strips of grass turf to people who want a lawn in a hurry. [...] Suppose I say to you, “Cut half an acre of grass for this customer”; I might mean not that you should mow it, but that you should slice it into strips as you could cut a cake or a loaf of bread. (Searle 1980, p. 224–5)

Moreover, he points out that there is a difference to be drawn between what he calls background assumptions and the special context of utterance for a given utterance. While background assumptions are involved in determining a sentence’s literal meaning or truth conditions, the context in which a sentence is uttered helps the hearer to decide on whether a speaker intended her utterance to be taken literally or non-literally. However, since Searle does not explicitly define what constitutes background assumptions and what is part of the context of an utterance, the question arises whether this differentiation really is necessary. From the examples Searle uses to defend his view of what constitutes literal meaning, it could be argued that the background assumptions necessary for determining the literal meaning of a sentence are in fact part of the specific context in which an utterance takes place. What Searle obviously means by background assumptions are certain aspects of knowledge that we have, namely those aspects which are relevant in the particular utterance situation. Thus, one could also assume that depending on the situation speakers and hearers find themselves in that situation will make certain aspects of knowledge they have more prominent (or salient). Those aspects, then, constitute what Searle calls background assumptions in the sense that speakers and hearers are presumably normally not aware of basing their utterances and interpretations on such assumptions. As their name implies, background assumptions are in the background; they form the basis from which speakers formulate their utterances and hearers intepret them. Thus, background assumptions depend on the particular context of utterance and therefore can be said to form part of the contextual information used in interpreting. Thus, in order to disambiguate whether the expression cut is used with the meaning as in ‘cut the grass’ or with the meaning as in ‘cut the cake’, the hearer needs to take into account contextual information. That is, even if the background of the utterance is such as Searle gives it, the hearer would still have to decide that the reading of cut as in ‘cut the cake’ is the one the speaker intended in that situation. From what has been said about salience of meaning above, it is of course very likely that the particular utterance situation will speed up the hearer’s unconscious decision.

Literal Meaning and Context-Independence

21

Furthermore, Searle’s argument that there are possible circumstances in which cut in ‘cut the grass’ may be interpreted as in ‘cut the cake’ actually does not constitute an argument against the assumption that the interpretation of cut is influenced by the intrasentential context. That is, one could assume that the co-occurrence of particular lexical items does help the hearer to narrow down the possible sentence meaning. However, this influence on the interpretation might have a default character. Thus, it only applies where the particular contextual conditions do not prevent it from applying. In the context Searle supplies, the interpretation of cut as in ‘cut the grass’ is rendered less likely as being the intended reading, than the reading as in ‘cut the cake’. As Searle argues, this reading of cut in the given context does not seem to be a non-literal reading of cut, since, intuitively at least, it does not seem to be derived from some clear basic, underlying meaning. However, assuming that the reading only comes about, or is interpretable as intended, in a particular context of utterance, suggests once again that literal meaning should not be taken to be a phenomenon of contextindependent sentence meaning. If that is the case, then the concept of literal meaning is not applicable at Searle’s level of sentence meaning, but rather at some context-dependent level of meaning. Be that as it may, Searle still assumes that literal meaning is the basis for any non-literal meaning. It is speakers who may use some expression or other non-literally. Thus, non-literal meanings have to be intended and should be expected to be consciously recognisable as such. That is, speakers should have no difficulty identifying some reading as being nonliteral, as they have to intentionally use some expression ‘deviantly’ in order for that expression to get interpreted non-literally. Although Searle thus argues against the view according to which literal meaning is determined by the linguistic system alone, he does not want to deny that sentences in fact do have literal meanings. ‘Literal meaning, though relative, is still literal meaning.’ (Searle 1978, cited from Searle 1979, p. 132). However, he applies the concept of literal meaning to those cases in which the speaker means what she says, contrasting them with those cases in which the speaker means more, or something different from what she said (e.g., cases of irony, conversational implicature or indirect speech acts). Thus, although Searle identified literal meaning as belonging to the level of sentence meaning, actually the differentiation between literal and non-literal meaning seems to hold at the level of utterances. However, he also argues that literal meaning is a relative notion. That is, it is rather likely that what we take to be the literal meaning of the utterance of some sentence will depend heavily on the specific circumstances in which the utterance of that sentence takes place. This is in agreement with uses of the term literal meaning where it refers to an utterance’s meaning, in contradistinction to the meaning intended by the speaker of that utterance.

22

Against the Standard Notions of Literal Meaning and Non-literal Meaning

Similarly to Searle, Bierwisch (1979, 1983) assumes that what is called the literal meaning of an utterance of some (simple or complex) expression is not identical to the linguistically determined meaning of that expression. Thus, Bierwisch also places literal meaning at a level of meaning that is no longer independent of context, namely the level of utterance meaning. Utterance meaning is the meaning an utterance token of an expression has when it is used in a context. The utterance meaning can be equivalent to the utterance token’s literal meaning, but it does not necessarily have to be. Therefore, Bierwisch differentiates the literal meaning of an expression from its utterance meaning. Crucially, an utterance token of an expression can only have literal (or, for that matter, non-literal) meaning in a context. With respect to this assumption, Bierwisch and Searle hold similar views. That is, what is called the literal meaning of an expression is not determined language internally, rather, it is dependent on certain background assumptions (Searle) or a particular context of utterance (Bierwisch). Thus, literal meaning is a special case of utterance meaning. A consequence of such a view is the assumption that the lexical semantic representations of simple expressions do not encode what we take to be their literal meaning. What has been said sofar, corroborates a suspicion expressed at the beginning of this section. That is, one has to ask whether the particular standard characterisation of literal meaning might not be very much influenced by our characterisation of the field of semantics. Thus, consider again that, traditionally, semantics takes as its starting point sentence meaning of which it assumes that it is both context-independent as well as propositional. Because it is taken to be context-independent, it seems to be what the sentence literally expresses. However, we saw that, for a large number of sentences, it cannot be said that they express propositions, unless contextual information is first taken into account (e.g., cases of reference resolution, fixing of indexicals and disambiguation)). That is, sentence meaning and ‘propositional content of an utterance’ are not equivalent. Furthermore, as we saw from Searle’s remarks, actually, what we take to be the literal meaning of a complex expression, is dependent on the context in which that expression is uttered. That is, essentially, literal meaning seems to be contextdependent after all. Thus, it seems reasonable to posit a partial new characterisation of the terms sentence meaning and literal meaning (with respect to sentences). Whereas the former is the meaning of a certain type of complex expression and characterised by the facts that it is compositional, context-independent and (more often than not) sub-propositional, the latter is a certain type of meaning a sentence may have when used in a context that allows a literal interpretation and in which that sentence expresses a full proposition. What this characterisation of literal meaning suggests is that whatever differentiates between literal meaning and non-literal

Literal Meaning and Context-Independence

23

meaning cannot be the criterion of context-(in)dependence. Moreover, this characterisation of literal meaning (with respect to sentences) makes it equivalent to a particular type of proposition. Thus, a new question arises, namely, how this particular type of proposition is characterised. That is, which conditions does a proposition have to fulfil for it to be literal in meaning? This is a question which is very close to the core of the discussion around the semantics/pragmatics distinction and we will come back to it in the following chapters. meaning

context-independent

lexical m.

sentence m.

under s pec i f i ed

context-dependent

...

l i t er al

...

...

nonl i t er al

Fig. 2.5: Revised differentiation of types of meaning

2.1.3 Literal Meaning as Primary to Non-literal Meaning?

The traditional view of the interpretation process assumes that an utterance’s literal meaning is always activated first. Potential non-literal meanings only get activated as a result of the literal meaning’s not fitting in the respective context. This view, of course, is based on the traditional assumption about literal meaning being context-independent. Since this allows an interpretation of the literal meaning of an utterance without taking recourse to contextual information, it will be computed automatically. If afterwards it becomes apparent that the literal meaning does not fit the contextual circumstances, a reinterpretation will take place, resulting in a non-literal interpretation of the utterance (call this the standard pragmatic view). This is essentially how Grice must have viewed the relation of literal meaning and non-literal meaning, since he described different kinds of non-literal meaning, such as irony and metapor, as being conversational implicatures, that is, inferences that require a prior recovery of what is said by an utterance. Of this latter level of meaning, Grice said that it is very closely connected to the conventional meaning of the words or the sentence uttered by the speaker. Specifying on this characterisation, what is said has been taken to be the fully propositional

24

Against the Standard Notions of Literal Meaning and Non-literal Meaning

semantic form of the utterance resulting from the processes of disambiguation, reference resolution and fixing of indexicals), where these processes already are of pragmatic nature since they involve contextual information (see figure 2.6).7 Semant i c s

lexical meaning s emant i cc ompos i t i on

sentence meaning Pr agmat i c s

r ef er enc er es ol ut i on f i x i ngi ndex i c al s di s ambi guat i on basis for further pragmatic inferences c onv er s at i onal i mpl i c at ur e s peec hac t s ...

does not involve any non-literal meaning

what is said

may involve non-literal meaning

what is meant

Fig. 2.6: (Non-)Literal meaning and Grice’s levels of what is said/what is meant

Thus, what is said, being a full proposition, provides the basis for further inferences about what the speaker actually meant with his utterance. That is, conversational implicatures are derived from the fact that the speaker said what he said in the particular way he did and with respect to a number of conversational maxims, which are taken to underlie human communication. Since what is said includes the conventional meanings of expressions, it is traditionally assumed to be the level at which literal meaning is expressed. And since Grice, similarly to Searle, viewed non-literal meaning such as irony or metapher, as an aspect of what the speaker meant – but not of sentence or utterance meaning – what is meant is the level of meaning at which such non-literal meaning aspects come into play. Thus, since for the recovery of what the speaker meant we first have to know what the speaker said, interpretation of the literal meaning of an utterance is prior to the interpretation of a potential non-literal meaning. Moreover, since non-literal meaning aspects such as irony or metaphor rest on the violation of a conversational maxim, they will only get derived if the literal interpretation, that is, what is said, cannot possibly be construed as the meaning intended by the speaker.

7 Note that this fact is independent of the assumption that what is said constitutes the level of meaning from which conversational implicatures are determined. That is, even if what is said is not viewed as wholly semantic, the standard pragmatic view could still hold in that what is said still forms the basis for drawing further inferences and thus has to be determined first.

Literal Meaning and Context-Independence

25

Interestingly, although arguing against viewing literal meaning as solely linguistically determined, Searle shares the view according to which literal meaning is prior to non-literal meaning. Thus, in his (1979) paper, Searle says of metaphorical and ironical utterances, that their respective interpretations are arrived at by going through the literal meaning of the sentences used to make the utterances. Thus, he seems to assume that although a sentence’s literal meaning can only be determined against particular background assumptions, in terms of the temporal progression of the interpretation process, literal meaning is a necessary intermediate step in the interpretation of non-literal meaning. Fortunately, with the methods developed in psycholinguistics, the assumption of the primacy of literal meaning as an assumption about the operational sequence of the interpretation process has become empirically testable. And in fact, results of experiments employing different methods in examining the understanding of various types of non-literal meaning in comparison to literal meaning show that the standard pragmatic view makes the wrong predictions. That is, in terms of cognitive effort, the standard pragmatic view predicts that interpreting non-literal meaning should be cognitively more exacting than the interpretation of literal meaning. Given that reaction or reading times mirror the relative cognitive effort involved in interpreting an utterance, results such as the following suggest that the interpretation of non-literal meaning does not necessarily differ from that of literal meaning. Thus, Gibbs (1994) mentions an experiment (Ortony et al. 1978), where subjects were presented with sentences in a context that was either literal or metaphoric. The hypothesis that was tested in the experiment was that people may not have to analyse the literal interpretation of a metaphorical utterance before actually deriving the intended metaphorical reading. The hypothesis was confirmed. Thus, although subjects took longer to read metaphorical targets than literal ones in short contexts, in long contexts, there was no difference in reading times for the metaphorical and the literal target sentences. These results suggest that the richness of contextual information available during the interpretation of an utterance has an effect on how difficult it will be to give that utterance a non-literal interpretation. Another experiment showed that utterances may be interpreted non-literally although there are no conditions that trigger the failure of a literal interpretation, suggesting that people automatically apprehend the metaphorical meaning of an utterance (Glucksberg et al. 1982). The task was to judge sentences such as Some jobs are jails as to their literal truth. Thus, it was not necessary to seek a non-literal interpretation for the sentences, subjects only were asked for the literal truth of each sentence. Now, if for a nonliteral interpretation of a sentence a pragmatic triggering condition is required, sentences such as Some jobs are jails should simply be considered as false. If, however, people automatically interpret the metaphorical meaning of such sen-

26

Against the Standard Notions of Literal Meaning and Non-literal Meaning

tences, then the ‘false’ judgement for the literal reading of the sentence should be in conflict with the ‘true’ judgement for the non-literal reading of the sentence. And in fact, although subjects correctly judged sentences such as Some jobs are jails as literally false, if a metaphorical interpretation for the sentence in question was available, subjects took much longer to make that judgement. Thus, apparently, the metaphorical meanings were automatically interpreted, without the need for some pragmatic triggering condition (i.e. maxim violation). This suggests that the interpretation of non-literal meaning does not rely on the violation of some conversational maxims or principles. Moreover, it shows that, although context may facilitate the interpretation of an utterance as non-literal, it is not absolutely necessary. Note, however, that more can be said with respect to such examples as Some jobs are jails. Thus, although it is true that there does not seem to be a pragmatic triggering condition such as maxim-violation for a metaphoric interpretation, nevertheless, it can be argued that in cases such as this, there IS some triggering condition, either semantically or pragmatically induced. To repeat, the standard pragmatic view assumes the literal meaning of an utterance is interpreted first and only if this does not fit the contextual circumstances a non-literal interpretation is determined. The problem with sentences such as Some jobs are jails or, for that matter, The ham sandwich is sitting at table 7 is that it is not clear what their literal meaning should be (cp. Stern 2006). Thus, the longer reaction times measured in the experiment mentioned above might be due to the fact that whatever component is responsible for this stage in the interpretation process is having problems determining a literal interpretation for the sentences in question. This, in turn may be sufficient to trigger an alternative, non-literal interpretation (cp. Dascal 1987). Keysar (1989) takes up this criticism and shows that even in contexts were a particular sentence is understood as literally true, that sentence’s potential but false metaphorical meaning interfers nevertheless. Thus, subjects were asked to judge test sentences as true or strongly implied to be true after having read small texts. The texts consisted of two parts, were one part related to the literal interpretation (L) of the target sentence, rendering it either true (L+) or false (L-), and one part related to the metaphorical interpretation (M), again rendering this either true (M+) or false (M-). Thus, texts were (L+M+), (L-M+), (L+M-) or (L-M-). For example, one of the test sentences was Bob Jones is a magician. An example text for which that sentence is interpreted as literally true but metaphorically false (L+M-) is given below. Bob Jones is an expert at such stunts as sawing a woman in half and pulling rabbits out of hats. He earns his living travelling around the world with an expensive entourage of equip-

Literal Meaning and Context-Independence

27

ment and assistants. Although Bob tries to budget carefully, it seems to him that money just disappears into thin air. With such huge audiences, why doesn’t he ever break even? (Keysar 1989, p. 378)

The results show that subjects are quickest in responding after texts that rendered the target sentence both literally and metaphorically true (L+M+). Generally, they are quicker in responding in congruent contexts (i.e. L+M+ and L-M-) than in incongruent ones. Thus, as in (Glucksberg et al. 1982)’s experiment, subjects take longer judging literally false but metaphorically true sentences. Crucially, however, they also take longer judging literally true but metaphorically false sentences, indicating that even in such a situation the potential metaphorical interpretation of the sentence is computed. This result shows that the longer reaction time measured by (Glucksberg et al. 1982) for literally false but metaphorically true sentences is not due to the difficulty of determining a literal interpretation for those sentences in the first place, as similar results are achieved in situations in which the target sentence actually is literally true but metaphorically false. Summing this section up, the conclusion one can draw from examining different empirical studies is that it is not generally necessary to compute the complete literal meaning of an utterance before deriving that utterance’s intended non-literal meaning (cf. Gibbs 2002). Thus, it seems that the possible but nonintended literal interpretation of an utterance does not constitute a necessary intermediate step during the processing of that utterance’s intended non-literal interpretation. Moreover, contextual information seems to be an important factor in the non-literal interpretation of utterances. Thus, if the context in which an utterance takes place points towards a literal interpretation, the utterance will be interpreted literally. If, however, the context of an utterance points towards a nonliteral interpretation, the utterance will be interpreted non-literally. However, it seems that context is not always necessary for the non-literal interpretation of an utterance. Thus, Glucksberg et al. (1982) found that sentences such as Some jobs are jails are interpreted non-literally, even if there is no context given with which the literal meaning could be said to be conflicting. The criticism that in such cases what may underlie the longer time needed for judgements concerning the literal truth/falsity of such sentences is not a maxim-violation problem, but rather the problem of determining a literal meaning at all was refuted by Keysar (1989). He found that even in contexts were a particular sentence gets a straightforward literal interpretation, a potential but non-fitting metaphorical interpretation interfers leading to longer reaction times in subjects. Thus, non-literal interpretations seem to get considered without the need of a triggering condition. With such results as the above, it becomes questionable whether it is warranted to assume a difference in conventionality for literal meaning and non-literal meaning.

28

Against the Standard Notions of Literal Meaning and Non-literal Meaning

2.2 Non-literal Meaning and Conventionality Whereas, traditionally, literal meaning was taken to be semantic, that is, contextindependent, conventional and primary, non-literal meaning was characterised as pragmatic, that is, context-dependent, non-conventional, a feature of the intended speaker meaning arrived at by a reinterpretation process and thus secondary to literal meaning. Having in the last section mainly argued against the traditional characterisation of literal meaning as context-independent and thus primary, in this section, I will argue that non-literal meaning should not be viewed as necessarily non-conventional and thus secondary. Generally, I want to argue that conventionality should not be viewed as an ‘all-or-nothing’ property, but rather as being gradual.

2.2.1 Empirical Evidence

Since non-literal meaning usually is considered a feature of language use, it is not surprising that the term non-conventional often features in its characterisation, where its opposite conventional is mostly used in the sense of ‘stored in the lexicon’. However, it is not necessarily so that a particular instance of non-literal meaning may not be considered as conventional. Rather, given that conventionality is reflected by familiarity and relative ease of interpretation, cases of conventionalised non-literal meanings of expressions are quite numerous.8 As a consequence, context may not actually play such an important role when it comes to the interpretation of an expression as literal or non-literal. Moreover, and as already suggested in the last section, the claim that non-literal meaning is always the result of a reinterpretation of some earlier composed literal meaning can no longer be sustained. Thus, consider the claims Rachel Giora (1997, 1999, 2002, 2012a,b) makes in the context of her graded salience hypothesis. Although she does not question the general importance of contextual information for utterance interpretation, she argues that regardless of the richness of the context, it can be shown that in particular cases both the literal as well as the non-literal meaning of an expression may be initially activated, suggesting that in such cases, the two types of meaning have the same status. More specifically she argues that it is the no-

8 Having said this, one might wonder whether under such circumstances it is actually still possible and useful to differentiate literal from non-literal meaning. It is if one wants to capture the fact that non-literal meaning, even if conventionalised, is perceived as in some sense non-basic or derived.

Non-literal Meaning and Conventionality

29

tions of salient vs. non-salient meaning that play a decisive role for the process of interpreting utterances. The crucial idea is that salient meanings are always activated, whether they are actually intended or not. Thus, in her (1999) paper, Giora mentions a number of experiments which, again, show that the non-literal meaning of an expression is not necessarily activated after its literal meaning has been. For example, Giora and Fein (1999) showed that reading times for familiar metaphors9 were equal for literally as well as metaphorically biasing contexts. That is, for familiar metaphors, both their possible literal and non-literal meanings are activated during interpretation, regardless of whether they are uttered in literally or metaphorically biasing contexts. This is because both meanings are familiar (salient) as for instance for to break ones head. (5) a. In order to solve the math problem, the student broke her head. b. Because she was so careless when she jumped into the pool, the student broke her head. Thus, the experiments show that in cases of familiar metaphors literal meaning and non-literal meaning of an expression are computed in parallel. In contrast, for less familiar metaphors only their literal meaning is salient, thus, it should be activated prior to the less familiar metaphoric meaning. This hypothesis was indeed confirmed by longer reading times found for less familiar metaphors in metaphorically biasing contexts as in contrast to literally biasing contexts. Giora (1999) cites the following sample contexts and target sentences for a less or unfamiliar metaphor.10 (6) Mary: My husband is terribly annoyed by his new boss. Every day he comes home even more depressed than he had been the day before. Somehow, he cannot adjust himself to the new situation. Jane: Their bone density is not like ours. (7) Our granny had a fracture just from falling off a chair and was rushed to the hospital. I told my sister that I never had fractions falling off a chair. She explained to me about elders. She said: Their bone density is not like ours.

9 This is a somewhat sloppy use of the term metaphor. What is meant is that expressions where investigated that could either be interpreted metaphorically or literally and for which the metaphorical interpretation is rather common. 10 It should be mentioned that the original experiment was carried out in Hebrew. Unfortunately, neither Giora (1999) nor Giora and Fein (1999) give examples for any of the original material. Thus, the critical points I will make below concerning the material and the differentiation of idiom from metaphor could just be an artefact of the English translation. However, as an Hebrew informant tells me, the criticism also holds when considering the Hebrew equivalents.

30

Against the Standard Notions of Literal Meaning and Non-literal Meaning

Note, however, that at least for the example cited, the longer reading time may also at least partly result from a further difficulty encountered during the interpretation process. Thus, whereas, arguably, in (5a) and (5b) the binding of the anaphoric expression her is equally difficult, the case is different for their in (6) and (7). Whereas in (7) their can be bound to the overt antecedent elders in the preceding sentence, there is no overt antecedent for their in the preceding sentence in (6). Thus, the longer reading time may reflect the greater difficulty in binding the anaphoric expression their to an appropriate, not explicitly expressed antecedent. A similar example of metaphoric material taken from Giora and Fein (1999) suggests that this is a general problem of the testing material used to investigate differences in the interpretation of metaphoric vs. literal interpretations. (8) The Saturday night party went on for hours. Drinks were poured, and we danced all night. We were probably less than considerate when, the next evening, we called on our friends who had been partying with us. When they opened the door, we realised: Only then did they wake up. (9) A bloody war has been going on in central Europe for a few years. Thousands of innocent women, men and children got massacred, and no one budged or lifted a finger. At last, a decision was made to intervene in the fights. Only then did they wake up. As in the example before, in (8) the anaphoric expression they in the last sentence can be bound to an appropriate antecedent in the preceding sentence, namely our friends. This, however is not the case for they in (9), where there, once again, does not seem to be an appropriate, overt antecedent given. An attendant, general problem is the fact that the context preceding the target sentence may also prime the meaning intended in the target sentence. Thus, the fact that in (7) in the sentences preceding the target sentence, such expressions as fracture and fraction are used, facilitates a literal interpretation of bone density in the target sentence. There is no such priming effect for the metaphoric interpretation of bone density in (6). The results of another experiment employing a word fragment completion task (Giora and Fein 1999) also suggests that the interpretation process varies with the familiarity of the metaphor processed. In this experiment, subjects were instructed to read little texts, the last sentence of which was the target sentence, involving a metaphor. Depending on the nature of the preceding context, subjects were intented to be biased either towards a literal interpretation or a metaphorical interpretation of the target sentence. Their task then was to complete two fragmented words, one of which was related to the literal (or rather compositional) interpretation of the target sentence, and one which related to the metaphorical interpretation. The results showed that in the case of familiar metaphors -

Non-literal Meaning and Conventionality

31

regardless of the bias induced by the preceding context - both the compositional as well as the metaphorical meaning were activated. In contrast, for less familiar metaphors, there was near to none activation of the metaphorical meaning in the literally biasing context, and more activation of the literal meaning than the metaphoric meaning in the metaphorically biasing context. Giora and Fein (1999) also tested the activation of meaning during the interpretation of familiar and less or unfamiliar idioms. The assumption was that for familiar idioms, their idiomatic meaning is more salient than the possible literal meaning. Thus, the idiomatic meaning should get activated both in idiomatically biasing as well as literally biasing contexts, whereas the literal meaning should only get activated in literally biasing contexts. As before, the task was to complete fragmented words with the first word that came to mind after reading an either literally or idiomatically biasing context with the last sentence being the target (see example below). This time, however, subjects were asked to complete only one of the two given fragmentations. Again, the results of the experiment confirmed this hypothesis. (10) a. He told me the whole story on one leg. b. In the zoo, I saw a stork standing on one leg. Thus, what the results of the different experiments mentioned in this and the preceding section show is that literal meaning is not necessarily processed first in interpretation. Moreover, once a non-literal meaning of an expression has grown familiar, it can be activated in parallel to the literal meaning of that expression, suggesting that non-literal meaning should not be seen as a type of meaning that belongs to a level of meaning, necessarily different from the one to which literal meaning belongs. Thus, the results from psycholinguistic experiments reinforce the need to rethink the view of the relation between literal meaning and non-literal meaning. That is, although historically and with respect to new or ad hoc instances of figurative meaning, the latter is derived from literal meaning by certain processes, such as metaphorical use, synchronically speaking this fact may no longer play a role. More specifically, in terms of the operational sequence of the interpretation process, the fact that the non-literal meaning of an expression may be activated in parallel to its literal meaning rather suggests that they are more similar in nature than traditionally assumed. However, the results of experiments such as the above cannot be used as evidence for one view or the other of literal meaning and non-literal meaning without caution, as there are a number of issues that need careful consideration. Generally, a lot of psycholinguistic studies are interested in phenomena such as metaphor, idioms and irony, none of which are uncontroversial. That is, there is a lot of debate concerning the appropriate characterisation of these phenomena.

32

Against the Standard Notions of Literal Meaning and Non-literal Meaning

Particularly with respect to metaphor and idioms, it does not seem to be trivial differentiating the one from the other, especially since one way for idioms to get the particular meaning they have is through a metaphorical use of the phrase in question. Thus, looking again at the examples for a metaphor and an idiom from Giora given above, I would content that it is not intuitively clear by which criteria one is a metaphor and the other an idiom. (11) to break ones head (12) on one leg Moreover, if metaphor is taken to differ from idiom by the fact that the metaphoric meaning is related to a single simple expression rather than to a phrase, the question arises which expression in to break ones head carries this metaphoric meaning.11 However, judging from the examples Giora gives for metaphors, she does not seem to assume that metaphoric meanings are related to a single expression, which makes the question, what, then in her view differentiates between idiom and metaphor even more pressing. Furthermore, Giora’s use of the terms literal meaning and non-literal meaning with respect to idioms is rather confusing. It is generally assumed that idioms are coded in the lexicon as a whole form associated with the respective (more or less) idiomatic (or in Giora’s sense non-literal) meaning. This coded meaning is usually taken to be the literal meaning of idioms, simply because the idiom is not taken to be a decomposable phrase, but as a whole. Idioms are usually not conceived of as having literal meaning in Giora’s sense. It is the phrasal, compositional counterparts of idioms that can be said to have literal meaning. The problem is, that it is not totally clear how selection or activation in cases where idioms are involved works. That is, when a sentence such as He told me the whole story on one leg is read or heard, the question is at which point the idiomatic meaning is activated. That is, how many segments must the addressee have heard in order to activate the idiomatic meaning? It seems rather plausible then, that an addressee will initially assume that he is dealing with a compositional phrase and then revise this assumption as soon as it becomes clear that the phrase in question is one that is coded in the lexicon as such. That is, from the phonological surface form only it is not obvious whether a specific phrase is used with a compositional meaning or with an idiomatic meaning. In this sense idioms and their compositional counterparts are similar to homonyms. They have the same surface graphical or phonological

11 One would probably be inclined to say that break carries the metaphoric meaning, however, then the question arises of whether this reading of break can be described without referring to the expression head.

Non-literal Meaning and Conventionality

33

form, but differ in their meaning and, additionally, in their complexity. Given this assumption, of course, one would expect that both meanings should be equally activated, which, according to Giora and Fein (1999), they are not. The results in that study show that for familiar idioms, in an idiomatically biasing context, the compositional meaning is only marginally activated. However, since it does not seem clear on which basis Giora differentiates idioms from metaphors, the results cannot really be compared, since some of the example expressions that were classified as metaphors may quite as well be classified as idioms. Moreover, one cannot rule out that the results for (familiar) metaphors and idioms in the two experiments only differ due to the difference in testing procedure. Thus, recall that in the experiment on metaphors, subjects were asked to complete both of the two given fragmented words, whereas in the experiment on idioms, subjects only had to complete one of the two given words. The fact that for familiar idioms, the number of activated idiomatic meanings is so much higher than the number of activated literal meanings probably is due to the fact that the idiomatic meaning is the most salient, however, for all we know the possible literal meanings might be activated to some degree as well, but this simply isn’t captured by the design. In other words, had subjects been asked to complete both given words, the difference between the extent of activation of the idiomatic and literal meanings might have been smaller and thus more similar to the difference recorded for familiar metaphors. In addition, since the experiment involved a task that measures the activation of concepts after the actual interpretation process has already taken place, the results do not tell us anything about early, initial processes. In other words, for all we know, an idiom’s literal meaning (i.e., the compositional meaning of the idiom’s compositional counterpart) may still be activated first during the interpretation process, but may be suppressed as soon as it is clear that it does not fit the context. The degree to which such suppression takes place may have a significant impact on whether the initially activated literal meaning is still active enough to get accessed by the time the task is being solved.12 Generally, Giora (and Feins)’s results do not tell us what EXACTLY is going on during the interpretation process.13 Thus, Giora and Fein (1999) used a word fragment completion task, which, it can be argued, is only sensitive to the state of activation at a time when the interpretation process has already finished. Although reading times may be better reflective of difficulties encountered during the interpretation process, it still is not clear what, or which steps, exactly are reflected by

12 Cp. Cacciari and Tabossi (1988). 13 See also Dascal (1989), who also questions the appropriateness of assuming that reaction times reflect what happens early on during interpretation.

34

Against the Standard Notions of Literal Meaning and Non-literal Meaning

that longer reading time. Vice versa, similar reading times do not tell us anything either of what exactly is going on during the interpretation process and whether what we assume to be literal meaning and non-literal meaning are processed similarly. A study by Laurent et al. (2006) – testing the graded salience hypothesis with respect to the activation of meaning for idioms using the EEG methodolgy – could in principle have tapped into early phases of the interpretation process, however, the authors only recorded EEGs during presentation of the last word of the respective idioms and the target words following that. Thus, their results do not reflect initial processes either. However, although the results of such experiments do not clearly contribute to disentangling the notions of literal meaning and non-literal meaning, at least they do not provide evidence that the traditional view of the operational sequence of the interpretation process is accurate. On a more positive note, ignoring the problematic aspects of such experiments for the moment, it nevertheless seems significant that all these studies point towards the same conclusion, namely that literal meaning and non-literal meaning are not so different in kind as traditionally assumed. In fact, Mashal et al. (2008)’s study provides evidence for differential processing of salient vs. non-salient meanings, rather than literal vs. non-literal meanings. Although this study, using fMRI, cannot make any claims concerning the time-course of activation of salient vs. non-salient meanings during interpretation, it does show a difference in the brain areas involved in interpretation related to the salience of the interpreted meaning. Thus, Mashal et al. (2008) investigated the interpretation of highly familiar idioms as compared to their possible compositional interpretation and the interpretation of matched literal sentences. Recall that according to the graded salience hypothesis the (idiomatic) meaning of highly familiar idioms is more salient than the (‘literal’) meaning of their compositional counterparts. The results show that while for the interpretation of the idiomatic meaning of idioms and the interpretation of literal sentences – both assumed to be salient – there is no involvement of the right hemisphere (RH), interpreting the literal and less-salient compositional counterparts of highly salient idioms does involve the RH. Thus, in this study, involvement of the RH does not correlate with literal/non-literal meaning of the sentence interpreted, but rather with the degree of salience of the meaning interpreted.

2.2.2 Theoretical Considerations

There are also some theoretical considerations supporting the view that literal meaning and non-literal meaning should not be seen as concepts definable in terms of dichotomies. In particular, literal meaning and non-literal meaning

Non-literal Meaning and Conventionality

35

should not be differentiated on the grounds that the former is conventional meaning as opposed to the latter, which is non-conventional meaning. Generally, one has to ask, what property exactly the concept of conventionality is supposed to pick out. That is, is an expression’s conventional meaning that which is coded in the lexicon, or is conventional meaning the reading with which a certain simple or complex expression is frequently used, regardless of whether that meaning is pragmatically derived or coded in the lexicon? In this context, the studies mentioned above, which showed that there are cases in which a non-literal meaning cannot be supressed, are interesting. Thus, the fact that for highly familiar metaphors their metaphoric meaning is activated even in contexts, where a literal interpretation is intended, lends credit to the assumption that such metaphoric meaning is as conventional as literal meaning. In addition to such empirical evidence, there are tendencies in theoretical linguistics, in particular with respect to meaning change, to view conventionality as a gradual concept instead of as a binary notion.14 This actually seems to be necessary, if one wants to explain how meaning change can take place at all, since it does not take place abruptly, but gradually. Thus, following Morgan (1978), Busse (1991) differentiates between conventions of language and conventions of language use, which he defines from a functional-pragmatic perspective.15 Thus, conventions of language are manifest in conventional language use, that is, a use of linguistic items that is generally and consistently conceived of as an intuitively and thoroughly ‘literal use’ by the members of a language community (Busse 1991, p. 55). In contrast, conventions of language use are defined as a costumary and established procedure of use of a particular linguistic expression for a consistent communicative purpose, where the use of that expression presupposes a still recognisable literal meaning as a necessary means of arriving at that purpose (Ibid.). Busse (1991) argues that different degrees of conventionality can be identified on the basis of a threefold distinction of levels of communicative use of language. At the first level, language is used ‘literally’, in that the members of the language community cannot detect any sort of ‘underlying meaning’ (i.e. a use of the conventions of language). The second level is the so-called ‘indirect’ or ‘deviating’ use, where that use might also already be conventionalised again in some sense (e.g. idioms

14 Also see Nunberg (1978, esp. ch. 4) for a characterisation of conventionality that is gradient. 15 Note that Busse also argues against an absolute differentiation between literal meaning and non-literal meaning, which conceals the fact that in most concrete cases, the dividing line cannot be drawn according to general, objective features. However, he also acknowledges that the differentiation is one that is made by normal speakers on an everyday basis. That is, it corresponds to some intuition about language and thus should not be ignored by theory.

36

Against the Standard Notions of Literal Meaning and Non-literal Meaning

c onv ent i ons ofl anguage c onv ent i ons ofl anguage us e unc onv ent i onal l anguageus e

conventional ised

or routine formulas), however, where most speakers may still recognise a ‘literal’ use of the expression in question. Thus, Busse acknowledges the fact that a particular ‘indirect’ meaning may also be conventionalised to some extent by differentiating between conventions of language and conventions of language use. On the third level of use, speakers use items from the first and second level or both, however, this use constitutes a communicative sense that is only comprehendable within a particular situation or context of utterance. Referring to the levels identified above, one can differentiate degrees of conventionality, according to whether speakers can still differentiate literal or ‘underlying’ meanings from a given communicative sense.

literal meaning

f ul l y mor e

idioms

l es s

non-literal meaning not particularised CIs

Fig. 2.7: Busse’s view of degrees of conventionality

Again following Morgan (1978), Busse (1991) recognises three elements of the conventions of language use: situation (here: context), purpose and means. Thus, the clearer the connection between communicative purpose and linguistic means to the communication partners, that is, the more obvious it is to the participants that the literal meaning differs from the intended meaning, the further removed from a convention of language a particular use is. It then depends on the epistemic status of the purpose-means-relation whether what we are dealing with is a context-dependent pragmatic inference or a form of language use that is already consolidated to a convention (of use). At the point where the relation between purpose and means becomes totally obscure, it may be reinterpreted as completely arbitrary, that is, at this point the convention of language use is reinterpreted as a convention of language (Busse 1991, p. 57). Thus, Busse (1991) shows that Grice’s theory of conversational implicature together with his theory of meaning can be utilised for the explanation of how meaning change proceeds. A conversational implicature is the result of a chain of inferences, which goes from the literal meaning, context, purpose and further assumptions to the actually intended meaning. The shorter this chain of inferences becomes, that is, the fewer neces-

Non-literal Meaning and Conventionality

37

sary ‘links’ it has, the more conventionalised the expression in question is. This may happen, e.g., if certain inferences are no longer necessary, because a particular use has already turned into an established procedure of use (Busse 1991, p. 58). Busse (1991) argues that meaning change starts at the point where speakers are no longer able to reconstruct all links within a chain of inferences. The becoming increasingly unnecessary of links within an inference chain then corresponds to increasing degrees of conventionality of the expression under consideration (cf. König 1988, Rolf 1995). However, as long as there is some transparency of purpose and linguistic means, we are still in the realm of what Busse calls convention of language use. Only at the point where the relation between purpose and means becomes absolutely ‘dark’, does the convention underlying the use of the expression in question become a convention of language. Thus, on Busse’s view conventionality and non-conventionality still go along with the differentiation of literal and non-literal meaning, however, he acknowledges the fact that there is a ‘grey area’, where meanings are to some extent conventionalised (as conventions of language use), but are not yet perceived of as literal meanings (thus, are not conventions of language). Thus, whereas Busse would still characterise literal meaning as fully conventional (being part of the conventions of language), non-literal meaning can be characterised as either being non-conventional or as being conventionalised to a lesser or greater extent. These differences in conventionality can be captured by assuming that the process of conventionalisation is a gradual one, leading to different grades of conventionality at different stages of that process. Looking, once again, at idioms shows that this characterisation still is problematic. As mentioned above, idioms are usually characterised as being coded in the lexicon as a whole, since their meaning is not (totally – cf. Nunberg et al. (1994)) compositionally determined. Thus, idioms may have compositional counterparts with meanings built up regularly from the meanings of their parts. Crucially, there are cases of idioms, where (at least synchronically) there is no transparent connection between the compositionally determined meaning and the meaning of the idiom, thus, making idioms and their compositional counterparts rather similar to homonyms (e.g. kick the bucket). As we saw, such idioms may still be characterised as having non-literal meaning16 (albeit their meaning is considered conventional), although this seems a rather irritating use of the term non-literal meaning. However, the reason for calling the meaning of an idiom nonliteral may simply be that speakers are, in principle, able to recognise that these expressions have counterparts that have compositional and thus ‘regular’ mean-

16 Cf. e.g. Giora (1999, 2002), Giora et al. (1997)

38

Against the Standard Notions of Literal Meaning and Non-literal Meaning

ings. In that case, the term non-literal is not only used for a meaning of some expression that is somehow ‘derived’ or based on some ‘underlying’ meaning, but also for a meaning of what ‘looks like’ a complex expression but is characterised by the fact that it does not have a (totally) compositional meaning. Thus, it is questionable whether normal speakers of English recognise a connection between the idiomatic and the compositional meaning of kick the bucket, in the sense that they take the idiomatic meaning to be derived from the compositional. One could argue, therefore, that idioms like kick the bucket are to be found on Busse’s first level of language use, the level where no ‘underlying’ meaning can be identified.

c onv ent i ons ofl anguage us e unc onv ent i onal l anguageus e

literal meaning conventional ised

c onv ent i ons ofl anguage

f ul l y

idioms

mor e l es s

non-literal meaning not particularised CIs

Fig. 2.8: Idioms in Busse’s framework

In this feature, idioms differ from metaphors in that with metaphors, the ‘underlying’ meaning always has to be transparent or accessible for us to be able to call the phenomenon in question metaphor. Thus, at the point when the underlying meaning of a metaphor is no longer transparent, the metaphor turns into a dead metaphor, effectively seizing to be a metaphor at all. However, that does not mean that certain types of metaphor may not be conventional. It simply means that in order to classify some phenomenon as a metaphor, recognising an ‘underlying’ meaning is necessary, whereas for idioms it is not.17, 18 Thus, it seems that whether we categorise a meaning of an expression as non-literal or literal, does not only have to do with whether we are consciously aware of some connection between the meaning in question and some underlying, basic meaning (as in metaphor), but also whether we recognise that a certain expression form can be used

17 Cp. Stern (2006, p. 250), who makes a similar point. 18 This is, of course, not to say that there are no idioms which are clearly perceived as ‘derived’ in their meaning.

Non-literal Meaning and Conventionality

39

c onv ent i ons ofl anguage c onv ent i ons ofl anguage us e unc onv ent i onal l anguageus e

conventional ised

with a different, compositional meaning (as with idioms). Thus, idioms are usually taken to be coded in the lexicon, since their meanings have to be learned and since they only superficially look like complex expressions with complex meanings associated with them. That is, in Busse’s terms, they are considered as being ‘conventions of language’, although according to Busse they should not be, since their meaning is perceived to be a so-called non-literal meaning – hence his classification of idioms belonging to the second level, the level of conventions of language use. However, there does not seem to be any independent reason why at least some idioms should not be treated similarly to expressions that are perceived as having no ‘underlying’ meaning. On the contrary, as mentioned above, from a processing perspective, the interpretation of idioms may not be any more ‘costly’ than the interpretation of ‘literal language’, where it seems that the level of conventionality of the idiomatic meaning plays a decisive role here. Thus, Giora recognises a type of idiom that she explicitly calls conventional idiom, where the crucial point is that in terms of activation their meanings are similar, if not privileged, to the literal (i.e., compositional) meanings of their compositional counterparts. That is, the meanings of highly salient idioms will be activated as early or even earlier than the meanings of their compositional counterparts, the latter being the case if the compositional meaning of some complex expression is actually non-conventional (in the sense of ‘not normally or often used’). Once again, if ‘earliness of activation’ reflects the level of conventionality of the meaning concerned, then there is no reason why one should assume that the meaning of such conventional idioms is less conventional than the meaning of their compositional counterparts.

f ul l y mor e

highly, less & nonsalient literal meaning

l es s

highly & less-salient non-literal meaning

not

non-salient non-literal meaning

Fig. 2.9: Salient meanings according to Busse’s characterisation

Generally, with respect to different types of what traditionally is called non-literal meanings such as metaphor or idioms, the experimental results differ with the grade of salience that these meanings have. However, according to Busse’s char-

40

Against the Standard Notions of Literal Meaning and Non-literal Meaning

acterisation of levels of conventionality, Giora’s salient and less-salient non-literal meanings belong to Busse’s conventions of language use, whereas the non-salient non-literal meanings are the result of what Busse calls an unconventional language use. As regards highly, less and non-salient literal meanings, according to Busse’s classification criteria, they are all part of the level of conventions of language (cp. fig. 2.9). In contrast, Giora does not actually make use of the traditional differentiation between literal and non-literal meaning, precisely because the conventional/non-conventional ‘divide’ seems to cut across that distinction. Thus, she assumes that regardless of whether some meaning would traditionally be classified as literal or non-literal, if that meaning is highly salient, it is reasonable to assume that this meaning is coded in the lexicon and thus, is part of the conventions of language. Moreover, Giora assumes not only of highly salient meanings, but also of less-salient meanings that they are part of the lexical semantics of the relevant expressions. That is, an expression’s salient and lesssalient meanings form part of its lexical entry, whereas possible, but non-salient readings are pragmatically derived. Thus, Giora’s differentiation between salient, less- and non-salient meanings differs from Busse’s assumptions in an interesting way. Busse assumes that only fully conventionalised meaning is lexical meaning (or a convention of language), whereas less conventional or non-conventional meanings are not lexical. The former may be considered as a convention of language use, but not of language. Giora, on the other hand, assumes that both highly and less-salient meanings are coded in the mental lexicon, and thus are part of the conventions of language, whereas non-salient meanings are not. In terms of the differentation of what is traditionally called literal and non-literal meaning this means that whereas Busse only allows for genuine literal, basic, underived meanings to be entered in the lexicon, Giora allows both literal as well as non-literal meanings in the lexicon, provided they are salient, where salience correlates strongly with conventionality.19 Moreover, she does not only assume of highly salient meanings, but also of more and less-salient meanings that they are entered in the lexicon. What is problematic about this assumption is that one has to ask, how the grade or level of conventionality of an expression’s meaning can actually be measured and

19 Giora takes conventionality to be only one graded property among others that influence an expression’s meaning’s salience, such as frequency of use. Nevertheless, conventionality seems to play an important role. That is, other things being equal, the more conventional an expression’s meaning is, the more salient it will be. Moreover, from the kind of conventionalisation process that Busse suggests, conventionality seems to be closely connected to the frequency of use of an expression with a particular reading. Thus, it can be argued that it is increased frequency of use that leads to a shortening of the inference chain ending in a conversational implicature.

Non-literal Meaning and Conventionality

41

c onv ent i ons ofl anguage

unc onv ent i onal l anguageus e

conventional i sed

thus allows one to identify and differentiate non-salient (and thus non-coded) from less or highly salient (and thus coded) meanings.20

f ul l y mor e

highly and less-salient meaning

l es s

not

non-salient meaning

literal and non-literal meaning

Fig. 2.10: Giora’s characterisation within Busse’s framework

Returning to the argument that literal meaning and non-literal meaning should not be characterised in terms of dichotomies: the discussion in the section on literal meaning above already suggested that one dichotomy between literal meaning and non-literal meaning cannot be sustained, namely that of context-independent vs. context-dependent meaning. Instead it was suggested that what traditionally is called literal meaning actually is not a complex expression’s context-independent, compositionally derived meaning, but rather a level of meaning that, similarly to non-literal meaning is context-dependent. From what has been said above concerning conventionality, it should have become clear that this property cannot be used either in the differentiation of literal from non-literal meaning, since, on the one hand, conventionality seems to be gradual and not binary, and on the other hand, there are instances of what is traditionally called non-literal meaning, which may be characterised as conventional to some greater or lesser degree. Thus, neither the dichotomy of conventional vs. non-conventional, nor that of context-independent vs. context-dependent can be used to differentiate literal meaning from non-literal meaning. As concerns the characterisation of non-literal meaning as always being the result of a reinterpretation process, similar points can be made. If reinterpretation processes are characterised as requiring the trigger of ‘non-compatibility of literal meaning with context’ to get started, then the process by which non-literal meanings are generated should not be called a reinterpretation process. That is, if non-literal meanings only get generated if the literal meaning of an expression does not fit the context, we would expect that

20 This point will be taken up in section 2.3.3.

42

Against the Standard Notions of Literal Meaning and Non-literal Meaning

non-literal meanings are never activated in contexts where the literal meaning of an expression is intended. As we saw above, this is not the case. Whatever triggers the activation of non-literal meaning, it does not necessarily have to be the non-compatibility of an expression’s literal meaning with the context.

2.3 Consequences for Lexical Meaning As mentioned above, formal semantics traditionally assumed that an expression’s lexical meaning can be derived from the meaning of a sentence the expression is part of via the principle of compositionality. Since sentence meaning was assumed to be context-independent, propositional and literal, lexical meaning is also taken to be context-independent and literal. However, in section 2.1 arguments were given against characterising sentence meaning as both context-independent as well as fully propositional. Rather, sentence meaning was characterised as context-independent, but not necessarily fully propositional. Moreover, arguments were given for not assuming that sentence meaning as such is literal, but rather that it is at some context-dependent and fully propositional level of utterance meaning that we might speak of a literal meaning, where it is not clear as yet how exactly this particular level of meaning is characterised. The crucial point here is that if sentence meaning is no longer considered to be literal but one still wants to assume that lexical meaning can be derived from sentence meaning via the principle of compositionality, lexical meaning can no longer be characterised as being literal either. In fact, I will argue below that the lexical meaning(s) of an expression should not be taken to be full readings at all. First, however, I will present some of the problematic data the semantic component of a grammar may or may not be required to be able to handle (2.3.1). What is important to note here is that the stance one takes in this question has consequences for the amount of lexically given semantic information assumed. I will then go on and discuss two general types of approaches to lexical meaning which differ mainly in how rich they assume the lexical semantic component to be (2.3.2). As we will see, approaches that do not assume underspecification of lexical meaning have quite a number of problems to face concerning the literal/nonliteral meaning distinction they assume, the idea of grades of conventionality of meanings and the roles of the principle of compositionality and context in semantic composition. Finally, I want to show that an approach that assumes underspecification for lexical meaning forms an appropriate basis for a view according to which both literal and non-literal meaning are context-dependent and more or less conventional (2.3.3).

Consequences for Lexical Meaning

43

2.3.1 Problematic Data

Traditionally, it is assumed that the lexical meaning of an expression captures those readings that are ‘basic’ in the sense that they do not depend on an actual utterance context. A potential problem with such an approach is that it leads to a high degree of (non-idiosyncratic) polysemy in the lexicon. Consider example (13). (13) a. The newspaper was on the table. b. The newspaper was censored, yesterday. c. The newspaper was founded three years ago. It seems intuitively obvious that the expression newspaper is understood differently in the individual sentences. Thus, in (13a) it refers to a physical object, in (13b) to a mental object and in (13c) to an institution. The difference in meaning is not due to a difference in the utterance context, which is completely ignored. Traditionally, one might therefore assume that the different readings of newspaper are all coded in that expression’s lexical entry and that newspaper has to be considered as polysemous. However, this type of variation in meaning is not characteristic of newspaper alone. That is, it is not an idiosyncratic property of that expression. Compare example (14) (originally due to Bierwisch 1983). (14) a. Die Schule spendete einen größeren Betrag. ‘The school donated a larger sum.’

b. Die Schule hat ein Flachdach. ‘The school has a flat roof.’

c. Die Schule macht ihm Spaß. ‘He enjoys school.’

d. Die Schule ist eine der Grundlagen der Zivilisation. ‘The school is one of the bases of civilisation.’

Thus, both newspaper and school (or Schule) have in common the ‘physical object’ and ‘institution’ reading. The latter reading allows for a metonymical interpretation as ‘person associated with institution’ of the NPs in (15a) and (15b), respectively. In this respect newspaper differs from book which, although it does have the ‘physical object’ and ‘mental object’ readings, does not have an ‘institution’

44

Against the Standard Notions of Literal Meaning and Non-literal Meaning

reading and therefore does not allow the metonymical interpretation ‘institutionfor-person’.21 (15) a. The newspaper called. b. Die Schule hat angerufen. ‘The school called.’

(16) a. The book was on the table. b. The book was censored, yesterday. c. ??The book was founded three years ago. d. ??The book called. The different readings given for the individual nouns in (13), (14) and (16a-16b) are a good example of what is traditionally called systematic polysemy. The individual readings of the respective noun are clearly delimitable, yet at the same time relate to one another. Also, all the readings seem to have the same status: they all intuitively seem to be basic, or literal, in the sense that no underlying meaning from which they are derived is consciously detectable. Moreover, they are not characteristic for only one expression, but rather can be found across sets of expressions. Thus, with respect to the lexical entries of the nouns above, it is questionable, whether one should assume that all of the relevant readings are listed as possible literal meanings of those nouns (cf. Nunberg 1979). Note also that the variation in meaning exemplified above seems to arise from the intrasentential context in which the respective expressions occur. Thus, the differences between the readings in (13a-13c), (14a-14d) and (16a-16b) are obvious simply from comparing the respective sentences with one another. It is arguable that sentential context, although being a special type of context in that it does not depend on a particular utterance situation, still is a type of context and that, thus, the readings of the expressions in question is not totally context-free, as it depends on the sentential contexts in which they occur. In fact, one actually needs to use fully sentential examples to make obvious that there are differences in the understanding of the expressions concerned. Moreover, even though the relevant examples are presented out of a particular context, as pointed out above, it is reasonable to assume that hearers or readers always interpret linguistic expressions against a number of background assumptions, which may be taken to have a default character.

21 This is not to say that there is no metonymical interpretation possible in principle. Thus, in a suitable context The book in (16d) could, e.g., be interpreted as ‘product-for-producer’.

Consequences for Lexical Meaning

45

In contrast to the related readings in case of systematic polysemy, strictly speaking, the metonymical readings for (15a) and (15b) above are not ‘basic’. That is, our interpretation of such examples relies on general knowledge we have regarding institutions, namely, that people (e.g., as employees) are associated with them. One might wonder, whether this reading is also part of the lexical meaning of the respective nouns, as it can be understood in the absence of a particular utterance context. However, as suggested before, the metonymical interpretation ‘person associated with institution’ is generally available for all lexical items which have an ‘institution’ reading, rather than being an idiosyncratic lexical feature of individual lexical items. Thus, the NPs in example (17) with such nouns as university, hospital, police get a similar metonymical interpretation. (17) The university/hospital/police called. In addition, more often than not context does play an important role in determining the exact interpretation of a metonymically used expression.22 (18) a. The ham sandwich in the corner wants some more coffee. b. Plato is on the top shelf next to Russell. c. John got a dent in his left fender. Thus, in (18c), the proper noun John apparently has to be understood as referring to John’s car. In this case the interpretation of John as meaning ‘John’s car’ seems to be constrained by the sentential context, i.e., its co-occurrence with the expressions dent and fender. (18b) is a typical example for the metonymical interpretation of names of authors, which can then be used to refer to books written by those authors. However, again the actual context in which the sentence gets uttered plays an important role. Thus, in a suitable context the terms could be referring to the busts of the respective persons just as well. Similarly, depending on the actual context of utterance, ham sandwich in (18a) can be understood as, e.g. ‘ham-sandwich-orderer’, or ‘person-who-prepared-the-ham-sandwiches’. More specifically, in examples like (18a), the sentential context already makes it clear that a shift in the meaning of the noun ham sandwich from ‘object’ to ‘person’ is necessary (especially due to the verb want). However, the exact interpretation or specification of the meaning (i.e., ‘person who prepared...’, ‘person who or-

22 The same, of course, holds true for particular cases of systematic polysemy. That is, for examples such as in (i) and (ii), the sentential context alone does not make it clear which particular reading of newspaper and school, respectively, is intended. (i) He is not interested in the newspaper. (ii) She paid particular attention to the school.

46

Against the Standard Notions of Literal Meaning and Non-literal Meaning

dered...’, ‘person who delivered...’, etc.) depends on the actual utterance context (and, arguably, the speaker’s intentions). Similarly, to understand an utterance of the sentence in (19a), recourse has to be taken to the actual context of utterance of that sentence. In addition, general background knowledge is necessary to provide potential interpretations for an utterance of the sentence. Thus, depending on the actual context of utterance, (19b – 19d) are possible readings for the sentence in (19a). (19) a. Joyce is difficult to understand. b. Joyce’s pronunciation is difficult to understand. c. Joyce’s behaviour is difficult to understand. d. Joyce’s books are difficult to understand. That is, in order to correctly interpret an utterance of (19a) in a given situation, the hearer needs to have some general knowledge concerning properties characteristic for the person named Joyce. Given that, it seems even less plausible that all the possible readings for a proper name such as Joyce are listed in that expressions’ lexical entry. Note that this example (as did Searle’s cut-example) shows that verbs have a variation potential with respect to their possible meanings as well. In this particular example, the meaning of understand seems to co-vary with the meaning assigned to the proper name Joyce. That is, understanding someone’s pronunciation, behaviour or books differs conceptually. Another example for such a variability of verb meaning is given in (20). (20) a. John opened the door. b. John opened the exhibition. c. John opened the file. d. John opened the parcel. e. John opened the store. Again, the processes which are named by the verb to open in (20) are intuitively conceived of as being conceptually different. In contrast to the readings of the term newspaper, which are clearly delimitable, this does not seem to be the case for the readings of to open or to understand. Although there is some sort of similarity between the readings for each expression, the relation between them does not seem to be of the kind that one would call systematic. Thus, Dölling and Zybatow (2007) call this particular type of meaning variation of verbs nonsystematic and characterise it as, in a sense, idiosyncratic for the respective verb that exhibits it. Nevertheless, similarly to the readings of systematic polysemous nouns, intuitively at least all the readings of open given above seem to be basic.

Consequences for Lexical Meaning

47

Again the question arises, whether one should assume that all these readings are therefore listed in the verb’s lexical entry making open – and a large amount of verbs generally – polysemous. As was the case for systematic polysemous nouns, the fact that the relevant readings of open are all basic would be masked by assuming that it only has one primary reading from which all other readings are derived. Moreover, there is a further type of meaning variation with verbs that is more similar to the systematic polysemy exhibited by nouns. Thus, consider the examples in (21). (21) a. John opened the door. b. The door opened. In (21a) open has a causative reading, whereas in (21b) it has an inchoative reading. Thus, (21b) describes a change of state, wheras (21a) describes the bringing about of this change of state. This particular type of meaning variation is systematic (cf. Dölling and Zybatow 2007) because, on the one hand, the different readings are related to one another and, on the other hand, it can be found with a number of verbs and, thus, is not idiosyncratic (cf. 22). (22) a. John melted the ice. b. The ice melted. Again, the question is how this particular variation should be treated with respect to the lexical entries of the verbs concerned. On the one hand, one wants to capture the fact that the different readings of the respective form are related to one another, on the other hand, one wants to capture that this particular variation in meaning is not idiosyncratic for one verb. Note that the possible non-systematic readings of a verb also differ depending on the valency with which it occurs (cf. 23). (23) a. Karl hat das Geschirr/ die Wäsche/ den Mund/ ... gespült. ‘Karl has washed the dishes/ the cloths/ his mouth ...’

b. Karl hat seiner Mutter das Geschirr/ die Wäsche/ ... gespült. ‘Karl has washed the dishes/ the cloths/ ... for his mother.’

c. (Mit Hinweis auf Geschirr:) Karl hat gerade gespült. ‘(With reference to dishes:) Karl has just washed (the dishes).’

Thus, in (23a) the verb spülen takes two NP arguments, whereas in (23b) it takes an additional benefactive NP argument. Although, in (23c) spülen only takes one

48

Against the Standard Notions of Literal Meaning and Non-literal Meaning

overt argument, it is implicitly understood that there has to be something which Karl washed, that is, the NP complement has to be present covertly in the semantic structure of the sentence. In any case, the processes that spülen denotes in the examples in (23) are conceived of as being conceptually different depending on whether and if, what kind of complement NP occurs. Furthermore, the interpretation of intransitive spülen seems fairly restricted to specific contexts. The last set of phenomena I want to mention here are cases of so-called complement coercion (e.g., 24b) and aspectual coercion (e.g., 26b).23 Thus, verbs like to begin and to finish may occur in two different syntactic constructions: either with an event-denoting VP-complement as in (24a) and (25a) or with an entitydenoting object NP as in (24b) and (25b). (24) a. Max began reading a novel. b. Max began a novel. (25) a. Nigel finished eating a hamburger. b. Nigel finished a hamburger. Note that in the latter case the respective sentences are nevertheless understood as predicating of the respective subjects that they began/finished some event involving the entity denoted by the object NP. However, this meaning aspect is not overtly expressed by any of the constituents in the respective sentences. Thus, it seems that the two verbs’ selectional restrictions actually are such that they require event-denoting complements. Since the sentences in (24b) and (25b) are grammatical, the usual assumption is that during the semantic composition of the meanings of these sentences a process intervenes that turns the entity-denoting object into an event description. Hence this process is called complement coercion. While in complement coercion the ontological sort of the complement is shifted (from entity to event), in aspectual coercion what is changed are the aspectual properties associated with a particular verb in order for it to fit with a given modification. Thus, consider the examples in (26). (26) a. The light flashed. b. The light flashed until dawn. Taken in isolation, the verb to flash is understood as denoting a punctual event. As such, a sentence including it as well as a durative modifier such as until dawn

23 Note that while coercion phenomena, on a first blush, may not seem to be relevant to the question of how much semantic information is lexically given, considering how the phenomena are explained in different approaches, it turns out that they are relevant after all.

Consequences for Lexical Meaning

49

should be ungrammatical. However, as (26b) shows, such sentences are grammatical, when understood as involving a repetition of the event denoted by the verb. A similar contrast can be seen in example (27). In (27a), the VP play a sonata denotes an event that already has an intrinsic boundary. If such a bounded event is modified by an adjunct that has the function of introducing a boundary for the event expressed, a conflict should arise. However, example (27b) is grammatical but again only under the interpretation that the subject was involved in the event denoted by the VP repeatedly. (27) a. Amelie played a sonata. b. Amelie played a sonata for three weeks. Thus, it seems that complement coercion and aspectual coercion are variants of one and the same general phenomenon. In both cases a process intervenes during the semantic composition of the relevant sentences and adds some meaning aspect. Put differently, the expectation is that the semantic composition of such sentences should crash as it apparently involves semantic mismatches. However, what seems to happen is that such impending (semantic) mismatches are circumvented by the application of special operators that mediate between the potentially mismatching elements and at the same time introduce additional meaning aspects. All in all, although this is only a very small set of data presented here, it suffices to indicate the kinds of problems one faces when developing a theory of lexical meaning and the role it plays in the overall interpretation process. In the next section, we will start to look at possible approaches to meaning in the lexicon and see how the data given above is treated, as well as how these approaches reflect the revised notions of literal meaning and non-literal meaning as proposed so far.

2.3.2 Approaches to Meaning in the Lexicon

Dölling and Zybatow (2007) identify three broad approaches to meaning in the lexicon: maximalist, intermediate and minimalist approaches. Generally, these approaches have in common the assumption that a lexical entry for an expression at least contain some phonetic (PHON), syntactic (SYN) and semantic (SEM)

50

Against the Standard Notions of Literal Meaning and Non-literal Meaning

information about that item.24 However, the approaches differ in how much ‘semantic material’ they assume to be given in the lexical entry of an expression, how this material is structured, how it is made use of in the actual interpretation of an utterance containing the relevant expression and how it interacts with the syntactic information given in the entry. Specifically with respect to the lexical entries of verbs, there is the question, whether information concerning argument structure and event structure are part of the lexical information given for a linguistic expression. In this section, I will be concerned with the maximalist and intermediate approaches, leaving the discussion of minimalist approaches to the next section.

2.3.2.1 The Maximalist Approach Maximalist approaches to meaning in the lexicon can be characterised by the following features. All possible literal meaning variants of an expression are listed in the lexicon. In addition to differences in the valency of a verb, the sortal restrictions for the individual meaning variants are stated as well. Thus, differences in valency and/or sortal restrictions result in separate lexical entries. Furthermore, the semantic relations holding between the individual meaning variants are stated. In the lexical entries of verbs, information about their argument structure is given as well as the respective verb’s particular event structure. An example of a maximalist approach is Engelberg (2000). For example, he gives the lexical entry, or rather entries, for the verb spülen (see figure 2.11). As one can see from the entry, not only differences in the valency of the verb, but also different sortal restrictions of its arguments are taken into account. As a result, Engelberg can account not only for the different readings of spülen in (28a-28c), but also for those in (28d) and (28e).

(28) a. Karl hat das Geschirr/ die Wäsche/ den Mund/ ... gespült. ‘Karl has washed the dishes/ the cloths/ his mouth ...’

b. Karl hat seiner Mutter das Geschirr/ die Wäsche/ ... gespült. ‘Karl has washed the dishes/ the cloths ... for his mother.’

c. (Mit Hinweis auf Geschirr:) Karl hat gerade gespült. ‘(With reference to dishes:) Karl has just cleaned (the dishes).’

24 Since here we are mainly interested in the interfaces between syntax, semantics and pragmatics, in the representation of lexical information the phonetic information will sometimes be neglected.

Consequences for Lexical Meaning

spülen1

SYN :

V, /NPakk /NPnom

SEM :

λyλxλe.

BP spulen1 : ¨

spülen2

spülen3

spülen4

spülen5

SPÜL 1 (x,

y, e)

 ∀xye[SPÜL1 (x, y, e) → GESCHIRR(y) ∨ WÄSCHE(y) ∨ ...]

SYN :

V, /NPakk /NPdat /NPnom

SEM :

λyλzλxλe.

SPÜL 2 (x,

y, z, e)

BP spulen2 : ¨

 ∀xyze[SPÜL2 (x, y, z, e) → BELEBT(x) ∨ BELEBT(z)]

BP spulen3 : ¨

 ∀xyze[SPÜL2 (x, y, z, e) → SPÜL1 (x, y, e)]

SYN :

V, /NPnom

SEM :

λxλe.

SPÜL 3 (x,

y, e)

BP spulen4 : ¨

 ∀xye[SPÜL3 (x, y, e) → GESCHIRR(y)]

BP spulen5 : ¨

 ∀xye[SPÜL3 (x, y, e) → SPÜL1 (x, y, e)]

SYN :

V, /NPnom

SEM :

λxλe.

SPÜL 4 (x,

y, e)

BP spulen6 : ¨

 ∀xye[SPÜL4 (x, y, e) → MENSCH(x)]

BP spulen7 : ¨

 ∀xye[SPÜL4 (x, y, e) → MUND(y)]

BP spulen8 : ¨

 ∀xye[SPÜL4 (x, y, e) → SPÜL1 (x, y, e)]

SYN :

V, /NPnom

SEM :

λxλe.

BP spulen9 : ¨

51

SPÜL 5 (x,

y, e)

 ∀xye[SPÜL5 (x, y, e) → WASCHMASCHINE(x)]

BP spulen10 : ¨

 ∀xye[SPÜL5 (x, y, e) → WÄSCHE(y)]

BP spulen11 : ¨

 ∀xye[SPÜL5 (x, y, e) → SPÜL1 (x, y, e)]

Fig. 2.11: Lexical entry for spülen according to Engelberg (2000)

d. (Beim Zahnarzt:) Karl hat gerade gespült4 . ‘(At the dentists:) Karl has just rinsed (his mouth).’

e. Die Waschmaschine hat gerade gespült5 . ‘The washing machine has just rinsed (the cloths).’

Note that for Engelberg, the sentences in (28c – 28d) are ambiguous. That is, only after having considered the context in which an utterance of the sentence Karl hat gerade gespült takes place, can the hearer disambiguate which literal reading of spülen was intended. However, if the context has to be taken into account in order to be able to decide which reading of spülen was intended, the principle of compositionality cannot be maintained. A possible counter-argument is to say that semantic composition actually generates all possible semantic forms for the sentence and only afterwards, during the process of disambiguation, the context is considered and those semantic forms that do not fit it are discarded. In this

52

Against the Standard Notions of Literal Meaning and Non-literal Meaning

case the principle of compositionality would be adhered to. However, considerations of economy of the lexicon militate against a maximalist organisation. This is especially so considering that this kind of ambiguity is characteristic for the majority of lexical items in a language (cf. the newspaper, open and cut examples). Furthermore, Engelberg apparently has not reached his aim of listing all possible literal readings for spülen. Thus, Dölling (2005) argues that a reading for spülen as in (29) is yet another one that would have to be listed in the lexicon. Note also that this reading cannot be subsumed under Engelberg’s spülen1 . Probably even more ’literal’ readings could be listed, if one were to do a corpus analysis to find attested examples (cf. Ruhl 1989, on data won through intuition). (29) (Mit Hinweis auf die Toilette:) Karl hat gerade gespült. For maximalist approaches, the process of semantic composition is essentially as assumed by traditional formal semantics. That is, the process follows the principle of compositionality and therefore, does not take into account any contextual information. Thus, for an utterance such as Karl hat gerade gespült, three semantic forms are generated by the process of semantic composition, as there are three lexical entries for spülen in its intransitive form. Thus, although the lexicon is very rich in what is usually considered world knowledge, this never really comes into play, since the disambiguation process that follows semantic composition simply checks, whether the information given in the individual semantic forms fits the information provided by the contextual conditions. In a sense, then, the information stated in the individual lexical entries for intransitive spülen is redundant. Moreover, if the lexical entry for an expression only lists its literal meaning(s), one would have to assume a Gricean interpretation process, where necessarily a literal reading for an utterance has to be derived first via semantic composition, which, in case of a non-fitting context, has to be reinterpreted. Assuming that all literal readings are listed in the lexicon for each lexical item would make the disambiguation process a very important part of the overall interpretation process and arguably, a rather extended one in terms of the time needed for that process to finish. If literal meanings are context-independent and coded in the lexicon, the process of semantic composition will determine as many possible sentence meanings as there are possible combinations of the literal meanings of the expressions concerned. Only in a second step, when contextual information is considered and the sentence meanings are enriched to the point that they are actually propositional, does the disambiguation process filter out all those propositions that do not fit the given context. Even assuming that all of this takes place during the short period of time that addressees usually only require to understand some utterance, one would still assume that of the time span needed for interpretation,

Consequences for Lexical Meaning

53

a relative large amount is spend disambiguating potential sentence meanings. However, the fact that potential non-literal meanings of some expression may be activated very early on during interpretation goes against such a view. Note also that an interesting question here is, which semantic form the reinterpretation process would be based on if during the first pragmatic interpretation, and due to a non-fitting context, none of the potential semantic forms derived during semantic composition could be chosen as fitting. Thus take once more the examples (15a) and (18), repeated below. (15a) The newspaper called. (18) a. The ham sandwich in the corner wants some more coffee. b. Plato is on the top shelf next to Russell. c. John got a dent in his left fender. Approaches that assume that semantic composition combines the literal meanings of simple expressions and thus produces the resulting complex expression’s literal meaning would be hard pressed to give the literal meaning of (18a) from which the metonymical reading is taken to be derived. This is especially so for approaches that assume that the sortal restrictions for lexical items form part of their lexical entry. In that case, arguably, the verb want would be specified as requiring an argument that has volition, which, literally, is not the case here. A possible way out would be to assume that ham sandwich is polysemous, one of its possible readings being ‘ham sandwich orderer’, and that this reading is listed in the lexicon. The question is whether one would like to assume such an entry and thus to make the process of metonymic shift a lexical one. As noted above, in view of the context-dependency of the results of an application of the process of metonymic shift, this is unlikely. Thus, as mentioned before, Plato in (18b) might be interpreted as ‘the person Plato’, ‘the book(s) written by Plato’ or possibly ’the bust of Plato’, depending on the particular context of an utterance of (18b). Regardless of how the question of which semantic form the reinterpretation process would be based on might be answered, the fact is that a maximalist approach that only allows literal readings of an expression to be listed in that expression’s lexical entry cannot account for cases of similar speed in, or even faster, interpretation of certain, conventionalised non-literal meanings. Thus, such a characterisation of meaning in the lexicon seems to assume the traditional view of a dichotomy of conventional vs. non-conventional meaning, for which, as we saw, there are arguments that it is not necessarily warranted. That is, a maximalist lexical entry does not reflect the fact that there are degrees of conventionality rather than just a differentiation between meanings that are conventional and such that are not.

54

Against the Standard Notions of Literal Meaning and Non-literal Meaning

2.3.2.2 The Intermediate Approach Intermediate approaches employ one of two general strategies. Either one specific meaning variant of an expression serves as the basis from which all other readings are deduced, or an abstract semantic representation provides the basis for all primary meaning variants of the expression. One example for an intermediate approach, using the first strategy is Pustejovsky (1991, 1995). With respect to the lexical entry for verbs, he states one specific meaning as their basic meaning, from which all others have to be deduced. Lexical entries are structured and involve a number of levels of representation: argument structure, event structure, qualia structure and lexical inheritance structure. For example, the lexical entry for the verb to bake looks as in figure 2.12.

⎡

bake

⎤

⎢ ⎢ ⎢EVENTSTR ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ARGSTR ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎣QUALIA



 ⎥ ⎥ = e1 : process ⎥ = ⎥ HEAD = e1 ⎥ ⎤ ⎡ ⎥  ⎥ ⎥ animate_ind ⎥⎥ ⎢ARG1 = 1 ⎥ ⎢ FORMAL = physobj ⎥ ⎥ ⎢ ⎥  = ⎢ ⎥⎥ ⎥⎥ ⎢ ⎦⎥ ⎣ARG2 = 2 mass ⎥ FORMAL = physobj ⎥ ⎥  ⎥  ⎥ state_change_lcp ⎦ = AGENTIVE = bake_act(e1 , 1 , 2 ) E1

Fig. 2.12: Lexical entry for to bake according to Pustejovsky (1995)

Thus, not only the syntactic order of lexical items, but also the semantic content of their entries may be decisive for how lexical items are combined. That is, the lexicon is actively involved in the semantic composition of sentences. For example, with the help of the information given in the qualia structure of bake, together with information from the lexical entry of the respective complement bake takes, problematic cases as in (30) can be explained. (30) a. Mary baked a cake. b. Mary baked a potato. Whereas in (30a) the verb to bake denotes the act of creating something, in (30b) it denotes a change of state. As can be seen from the lexical entry of bake, its basic meaning is that of denoting a change of state, thus, the reading that needs to be explained is that in (30a). Information about the internal argument, in this

Consequences for Lexical Meaning

55

case the noun cake, is used to shift the verb’s event type. Intuitively, an object such as a cake is brought about by the activity it is in composition with. This is expressed by the information stated in the qualia structure. Thus, the agentive quale of the noun cake makes reference to the process of baking.

⎡

⎤

cake

⎢ ⎢ ⎢ARGSTR ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢QUALIA ⎢ ⎣

 =

ARG 1 D -ARG 1

⎡

= =

CONST

⎢FORMAL ⎢ = ⎢ ⎣TELIC AGENTIVE

x:food_ind y:mass = = = =



y x eat(e2 , z, x) bake_act(e1 , w, y)

⎥ ⎥ ⎥ ⎥ ⎥ ⎤⎥ ⎥ ⎥ ⎥⎥ ⎥⎥ ⎥⎥ ⎦⎥ ⎦

Fig. 2.13: Lexical entry for cake according to Pustejovsky (1995)

A number of operations take place in combining the lexical structures of bake and (a) cake. Amongst them is the process of co-composition, which results in a semantic structure for bake a cake that incorporates aspects of both the constituents. Thus, in combining the lexical information of bake with that of cake, the resulting VP is interpreted like a genuine creation verb such as build. The resulting semantic form of the VP bake a cake looks like in figure 2.14. Thus, systematic (or as Pustejovsky calls it: logical) polysemy is explained on the level of expression meaning with the help of lexical semantic processes that compose larger semantic structures out of the lexical semantic information of the individual expressions. However, it is not clear what leads Pustejovsky to assume that the ‘change of state’ reading is the underlying, basic meaning of bake, especially since in the examples in (30), neither the one nor the other seem to have a derived meaning. Generally, an advantage of Pustejovsky (1991, 1995)’s approach is the fact that it provides a framework to model the role of sentential context in interpretation. Thus, the process of co-composition might be used to explain the phenomenon that the reading of a particular expression may be determined before the context of utterance is actually considered. That is because the lexical entries of linguistic expressions are relatively rich, consisting of different types of information, where that information interacts with the information coded in the lexical entries of expressions which occur in the same sentence or phrase. Note also, that Pustejovsky (1991, 1995)’s approach allows for a relative straightforward modelling of complement coercion in terms of the process of co-composition. A sentence like

56

⎡

Against the Standard Notions of Literal Meaning and Non-literal Meaning

bake a cake

⎢ ⎢ ⎢ ⎢ ⎢ ⎢EVENSTR ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ARGSTR ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢QUALIA ⎣

⎤ ⎥ ⎤ ⎥ e1 :process ⎥ ⎥ ⎥ ⎢E e2 :state ⎥ ⎥ ⎢ 2 ⎥ = ⎢ ⎥ ⎥

points to

{OPEN-DOOR[10], OPEN-WINDOW[9], OPEN-FILE[8], OPEN-EYES[8], . . .}

degree of accessibility

Fig. 2.18: A lexical entry adapted from Relevance Theory

2.3.3.4 Underspecification and Conventionality One other issue I would like to discuss is the relation of lexical meaning and the property of conventionality in approaches that assume a high degree of semantic underspecification. Thus, according to Saussure (1931), linguistic signs consist of two parts, a form and a meaning, where the relation between those two parts rests on a convention. That is, there is no natural connection between the meaning and the form of an expression. But, what do we mean by ‘convention’? (Lewis 1969) offers a formalised definition of convention. A regularity R in the behaviour of members of a poluation P when they are agents in a recurrent situation S is a convention iff it is true that, and it is common knowledge in P that, in almost any instance of S among members of P: 1. almost everyone conforms to R 2. almost everyone expects almost everyone else to conform to R 3. alomst everyone has approximately the same preference regarding all possible combinations of actions 4. almost everyone prefers that any one more conforms to R, on condition that almost everyone conforms to R 5. almost everyone would prefer that any one more conform to R’, on condition that almost everyone conform to R’

Consequences for Lexical Meaning

73

where R’ is some possible regularity in the behaviour of members of P in S, such that almost none in almost any instance of S among members of P could conform both to R’ and to R.’

Thus, conventions seem to have to do with ‘customariness’ within a group. That means that a speaker associates a certain expression form with a certain meaning, because a sufficiently large enough group of speakers associates that expression form with that meaning. Under such a view of convention, it seems inappropriate to characterise the abstract underspecified semantic form of an expression as being conventional in the intended sense, since SEMs are not actual readings of an expression. As mentioned above, there are a number of psycholinguistic studies which show that particular types of non-literal meaning may be interpreted as fast as literal meanings. In particular, it has been shown that certain non-literal meanings of expressions get activated even in contexts where it is obvious that a literal interpretation is intended. Moreover, the results differ with respect to whether the non-literal meanings are salient, less-salient or non-salient. According to Giora’s definition of salience, conventionality of a reading plays an important role. Moreover, I have already argued above, that conventionality should not be viewed as a binary feature and, thus, cannot be used to differentiate between literal and nonliteral meaning. Thus, the fact that under the assumption of underspecification of lexical meaning, conventionality in the intended sense cannot be taken to be a feature of lexical meaning, does not pose a problem. As the results of the empirical studies mentioned suggest, conventionality is a graded notion which applies not only to what traditionally is called literal meaning, but also to non-literal meaning, in some cases leading to a non-literal meaning actually being conventionally associated with some phonological form to a greater degree than a possible literal reading. Thus, approaches assuming highly underspecified semantic forms come very close to what Recanati (2004) calls meaning eliminativism (ME) and which he characterises as follows. ...according to ME, we do not need linguistic meanings even to serve as input to the construction process [...] the senses that are the word’s contributions to contents [of utterances] are constructed, but the construction can proceed without the help of conventional, contextindependent word meanings.’ (Recanati 2004, p. 147, my emphasis)

Although underspecification-approaches do assume some SEM-component in an expression’s lexical entry that SEM only serves as a ‘pointer’ to potential readings

74

Against the Standard Notions of Literal Meaning and Non-literal Meaning

of the expression.43 It does not constitute anything like the conventional meaning of an expression, as assumed under the traditional view of lexical semantics. Moreover, Recanati (2004) says of ME that it computes ‘directly the contextual sense which an expression assumes on a particular occasion of use on the basis of the contextual sense which that expression had on previous occasions of use...’ (Recanati 2004, p. 147). If, however, the contextually determined reading of an expression is computed on the basis of the readings of earlier uses of the same expression, there has to be some sort of connection between the expression and those readings: this is the radically underspecified SEM component.

2.3.3.5 Underspecification and Semantic Relations Traditionally, one of the goals of semantics was to describe and explain the meaning relations that were found to hold between individual expressions. However, it has been noted that this task is not at all easy to accomplish. Thus, under the traditional perspective of linguistic signs as consisting of a form and a meaning, first, individual lexical units (combinations of a distinctive sound shape and a distinctive meaning) which share the same phonological form, have to be classified as belonging to the same or different lexemes. Second, a particular lexeme which is characterised as being polysemous, thus, incorporating more than one lexical unit, may have several distinct meaning relations to several, distinct lexemes. What is generally problematic about the notion of semantic relations is that there is the danger that specifically linguistic knowledge is mixed up with general conceptual knowledge (cp. Fodor and Lepore 1998). Thus, the underlying assumption for the specification of meaning relations between expressions is that their intensions share some feature that allows their extensions to be related in a certain way. That is, the fact that the expressions human and woman are taken to be in a hyponymy relation is due to the observation that the denotation of woman forms a subset of the denotation of human. In order to explain this fact about the denotations of these expressions, it is assumed that their intensions, or meanings, are in a particular relationship as well, namely, the meaning of woman includes the meaning of human. However, the question is whether, when thinking about such relationships between denotations, it is admissible to link this

43 For an actual ME-approach, one that denies that words actually consist of both sound form and meaning but rather assumes that phonological form merely represents conceptual content (hence the name representational hypothesis), cp. Burton-Roberts (1994, 2000, 2007a), BurtonRoberts and Poole (2006), Chng (1999).

Consequences for Lexical Meaning

75

relation to a relation between their linguistic meanings. I think it is not. Rather, the relationship holding between the denotations of human and woman is due to the relation between the concepts HUMAN and WOMAN that we have and which, crucially, we have independently of the fact that we can use the phonological forms /hju:m@n/ and /wUm@n/, respectively, to express them. Thus, semantic relations are much more closely related to our general cognitive ability of categorisation than to linguistically relevant aspects of meaning – whatever those may be. Moreover, a speaker may be able to use a particular expression correctly without being aware of the relationships that hold between the concept that the expression he uses expresses and other concepts. That is, I might use the expression whale successfully to refer to an instance of the kind ‘whale’ without knowing that the denotation of the concept WHALE is a subset of the denotation of the concept MAMMAL. Thus, one might question Borg (2004b)’s claim that a semantic theory should make clear the relations that hold between individual complex expressions, for example, which inferential relations possibly hold between them. In fact, one consequence of assuming underspecification of the linguistic meanings of expressions is that semantics is no longer able to capture the various meaning relations an expression may enter into. In that case, how could semantics capture the inferential relations holding between elements of different domains that may be denoted by one and the same expression beyond only very general relations. More specifically, meaning relations may be taken to hold between full readings of expressions, but not between their abstract, underspecified semantic forms. However, from the arguments given above, it seems that one should not expect of semantics in general that it has anything to say concerning the relations holding between the concepts expressions may be used to express.

2.3.3.6 More Underspecification in the Lexicon The idea that the lexical information for some linguistic item does not include information on that item’s argument structure has not only been pursued for the semantic, but also for the syntactic component of such entries. Thus, Kratzer (1996) argues against the assumption that the argument that will occupy the subject position in a sentence is projected by the verb, rather, it comes into the structure by means of a functional head. Thus, for the sentence in (44), (Kratzer 1996, p. 121) gives the partial syntactic structure as in (2.19).

(44) Mittie feeds the dog. Thus, Mittie is base-generated in SpecVoiceP, whereas the head of VoiceP introduces the external argument. Note that Kratzer uses the ordered argument no-

76

Against the Standard Notions of Literal Meaning and Non-literal Meaning

VoiceP

DP Mittie

Voice’

VP

Voice Agent

DP the dog

V’ V feed

Fig. 2.19: Partial syntactic structure for Mittie feeds the dog

tation for argument structure, which does not clearly separate syntactic from semantic information. However, translating the facts into Dölling’s notation, the syntactic component of the lexical entry of feed would thus only specify that it is a verb and that it takes one NP (internal) argument. As Dölling (2005) notes, Borer (2004) goes even further in claiming that one should also separate the internal argument from its verb. Ignoring issues concerning the adequacy of such an approach for the moment, what this would result in is a syntactic component for feed, which would no longer include information on the number and kind of arguments that item takes. Thus, there are approaches in syntax also that reduce the amount of information required to be given in some linguistic expression’s lexical entry. In fact, there are attempts at radically reducing the role the traditional lexicon plays in the linguistic system. Thus, Marantz (1997) wants to dispense with the lexicon as traditionally conceived altogether. He assumes three distinct lists which to draw from in producing an utterance. The first of these, which he calls ‘narrow lexicon’, is the list of units that syntax operates with. This list contains a language’s atomic roots, as Marantz calls them, as well as the atomic bundles of grammatical features. What is interesting about the roots is that they are neu√ tral with respect to word category. That is, a root like DESTROY is neither a noun nor a verb; depending on the kind of syntactic structure which it is part of it may √ function as a verb or as a noun. Thus, merging a root like DESTROY with ‘D’ (i.e., a nominalising environment) places it in a nominal context (cf. Figure 2.20). In contrast, merging the root with a verbalising functional head (v-1 in 2.21) results in destroy functioning as a verb. True nominalisations, then, contain both a verbalising and a nominalising head (cf. Figure 2.22).

Consequences for Lexical Meaning

77

D √DESTROY

D

√DESTROY

the city

Fig. 2.20: E.g. the destruction of the city, the city’s destruction v-1

v-1

√DESTROY √DESTROY

the city

Fig. 2.21: E.g. (John)destroyed the city

What is interesting about this approach in light of the present concerns is the following. If lexical items (or rather roots) actually are characterised by the fact that they are category neutral, the syntactic information one would expect to find for them might be similarly underspecified as is the case for semantic information. Thus, the lexicon may not provide any distinct information on category membership of roots. However, and Marantz (1997) explicitly notes this, roots would need to include some type of information which ‘tells’ the syntax in which potential structures the root may occur. Thus, Marantz (1997) notes that a root √ √ like GROW differs from a root like DESTROY in that it does not imply an agent. √ Thus, GROW belongs to a different class (or possibly has different features) than √ √ DESTROY . Whereas GROW semantically denotes a change of state that is inter√ nally caused, DESTROY denotes a change of state that is not. More specifically, √ with DESTROY an external cause or agent is implied. However, assuming such a characterisation may not be unproblematic from a semantic point of view. Generally, it is unclear in which form this type of information enters into the lexical entries of the items in question. One plausible possibility is to assume that it is part of an expression’s semantic form. However, this is precisely what a minimalist approach wants to avoid. Another possibility is to assume some abstract ‘group’ feature for different classes of roots. Leaving the particulars to the side, it is still interesting to note the parallels between the minimalist approach and the syntactic/morphological approaches mentioned above, all of which argue for some underspecification of lexically provided information.

78

Against the Standard Notions of Literal Meaning and Non-literal Meaning

D D

v-1

v-1

√DESTROY √DESTROY

the city

Fig. 2.22: E.g. (John’s) destroying the city

2.3.3.7 Underspecification of Semantic Composition Similarly to Dölling (1997, 2005), Egg (1994, 2003, 2005) also assumes that lexically given semantic information for expressions is highly underspecified. However, in his approach, the result of semantic composition is not only underspecified due to the underspecified nature of the elements that it composes, but in addition, the resulting semantic form only consists of fragments and an indication of the dominance relations between them. Crucially, the fragments are not yet totally fixed with respect to the scope they take over one another. For an exemplification of Egg (2003, 2005)’s approach, consider again (32), repeated below.

(32) a. Max began reading every novel. b. Max began every novel.

□

Recall that whereas (32a) has two readings depending on the scope the universal quantifier takes, (32b) only has one reading, where the universal quantifier outscopes the verb to begin. The respective semantic forms of (32a) and (32b) in Egg’s approach resulting from semantic composition are given in figures (2.23) and (2.24).

A

λe' 'y . nov el '(y )

□

‖S‖ : λe. begi n' (max' , λz □ )(e)

(y )(z )

λy λx λe. r ead.(x ,y )(e) Fig. 2.23: The semantic form of Max began to read every novel.

□ (e' ' )

Consequences for Lexical Meaning

A

λe z . nov el '(z )

79

□ (e)

‖S‖ : □ (z )(max' ) begi n' Fig. 2.24: The semantic form of Max began every novel.

The squares indicate ‘holes’ in the semantic representation. These holes have to yet be specified. The dotted lines indicate dominance relations between the individual fragments. Thus, in (2.23), both the universal quantifier and the verb to begin dominate the verb to read, whereas the dominance relation between them is not yet specified. However, either specification is compatible with the relations already fixed in the semantic representation of (32a). This is different for (32b), where in its semantic representation (2.24), the dominance relation between the universal quantifier and the verb to begin is already fixed such that the former dominates and thus has scope over, the latter. I do not want to go into too much detail of Egg (2003, 2005)’s analysis here, but note that the difference in the semantic representations for the sentences (32a) and (32b) is due to differences in the syntax-semantics interface rules that determine how the respective sentence meanings are composed. Moreover, whereas in the case of the semantic representation of (32a), there is a so-called constructive solution of the underspecification in the semantic representation, there is no such solution in the case of (32b). A semantic representation has a constructive solution if the underspecified elements (‘holes’) may be resolved simply by integrating the material that is already present in the semantic representation itself. For example, in the semantic representation for (32a), the semantics of the verb to read can be inserted into the hole immediately dominating it. By functional application, the verb’s semantic type is converted from to , which makes it of the appropriate type for insertion either into the hole of the semantics of the verb to begin or that of the quantifier. Depending on into which of these holes the verb’s semantics is inserted, the resulting representation can then be inserted into the remaining hole, thus determining the scopal relations between the quantifier and the verb to begin.44 The problem with the semantic representation of (32b) is that a constructive solution is not possible, simply because the verb to begin as it occurs in that representation is not of the appropriate type for insertion into the hole immediately

44 Following Egg (2002) and Egg (2003), I ignore the topmost hole and its dominance relations here.

80

Against the Standard Notions of Literal Meaning and Non-literal Meaning

dominating it (it would have to be of type but actually is of type